init

.gitignore

@@ -0,0 +1,4 @@
+.Rproj.user
+.Rhistory
+.RData
+.Ruserdata

Event Data Analysis.Rmd

@@ -0,0 +1,343 @@
+---
+title: "Event Data Analysis"
+author: "Kunwu Lyu"
+date: "2024-06-26"
+output: pdf_document
+---
+
+```{r setup, include=FALSE}
+knitr::opts_chunk$set(echo = TRUE, 
+                      size = "small", 
+                      collapse = TRUE, 
+                      comment = NA, 
+                      warning = FALSE, 
+                      message = FALSE,
+                      error = TRUE) # change it to TRUE
+```
+
+
+## Loading Packages
+```{r}
+library(dplyr)
+library(tidyverse)
+library(stringr)
+library(lubridate)
+library(tidyr)
+library(forcats)
+library(ggplot2)
+library(viridis)
+library(viridisLite)
+library(purrr)
+library(gganimate)
+library(ggthemes)
+library(leaflet)
+library(patchwork)
+library(readr)
+library(janitor)
+library(plotly)
+```
+
+
+## Data Wrangling
+
+```{r}
+# Define the directory containing the files
+data_dir <- "data/"
+
+# Define the file codes
+file_codes <- c("F15", "W16", "S16", "F16", "W17", "S17", "F17", "W18", "S18", "F18", 
+                "W19", "S19", "F19", "W20", "S20", "F20", "W21", "S21", "F21", "W22", 
+                "S22", "F22", "W23", "S23", "F23", "W24", "S24")
+
+# Create a function to read a file given its code
+read_and_clean_event_file <- function(code) {
+  file_path <- paste0(data_dir, "2015-2024 Events Data - ", code, " - Event Data.csv")
+  data <- read_csv(file_path)
+  clean_names(data) %>% 
+    mutate(across(everything(), as.character)) %>% 
+    mutate(term = code)
+}
+
+term_to_year <- function(term) {
+  year <- as.numeric(str_sub(term, 2, 3))
+  season <- str_sub(term, 1, 1)
+  start_year <- if_else(season == "F", 2000 + year, 2000 + year - 1)
+  end_year <- start_year + 1
+  return(paste0(start_year, "-", end_year))
+}
+
+# Use purrr to read all files and store them in a named list
+event_data_list <- set_names(map(file_codes, read_and_clean_event_file), file_codes)
+
+combined_data <- reduce(event_data_list, full_join)
+
+combined_data_filtered <- combined_data %>%
+  filter(!is.na(what), what != "") %>% 
+  filter(what != "Choir & Jazz Rehearsal") %>% 
+  filter(what != "Jazz Rehearsal") %>% 
+  mutate(date = as.Date(ymd(date))) %>% 
+  mutate(support_level = if_else(support_level == "N" | support_level == "Y", "L", support_level)) %>% 
+  mutate(
+    department = str_replace_all(department, "WCC", "ODOA"),
+    department = str_replace_all(department, "MSUC", "MUSC"),
+    department = str_replace_all(department, "French Dept|French", "FREN"),
+    department = str_replace_all(department, "English", "ENGL"),
+    department = str_replace_all(department, "Pres. Office", "PRES"),
+    department = str_replace_all(department, "History", "HIST"),
+    department = str_replace_all(department, "THD", "THDA"),
+    department = str_replace_all(department, "Inclusion & Equity", "IEC"),
+    venue = str_replace_all(venue, "Skinner Chapel", "Chapel"),
+    department = str_replace_all(department, "/", " & "),
+    department = str_replace_all(department, ",", " &"),
+    venue = str_replace_all(venue, ",", " &"),
+    department = str_replace_all(department, "\\s+", " ")  # Remove extra spaces
+  ) %>% 
+  select(-wk) %>% 
+  mutate(department_type = case_when(
+      department == "MUSC" ~ "MUSC",
+      department == "ODOA" ~ "ODOA",
+      department == "CSA" ~ "CSA",
+      str_detect(department, "&") ~ "Collab",
+      TRUE ~ "Others"
+    )) %>% 
+  mutate(what = str_replace_all(what, "Jazz Ensemble Concert|Jazz Area Concert", "Jazz Concert"),
+         what = str_replace_all(what, "Symphony Band Concert", "Symphony Concert"),
+         what = str_replace_all(what, "Composition Recital", "Composition Showcase Recital"),
+         what = str_replace_all(what, "Harpichord", "Harpsichord"),
+         what = str_replace_all(what, "Emsemble", "Ensemble"),
+         what = str_replace_all(what, "Juest Cellin'", "Just Cellin'"),
+         what = str_replace_all(what, "Facutly|FACULTY|Mazariello", "Faculty")) %>% 
+  mutate(event_type = case_when(
+    str_detect(what, "GUEST|ODOA|Concert Series|SPCO") ~ "Guest",
+    str_detect(what, "Faculty") ~ "Faculty Recital",
+    str_detect(what, "Student|Senior|Junior|Piano Recital: |Johnson|Verma Jameson") ~ "Student Recital",
+    str_detect(what, "Studio Recital|Organ & Harpsichord|Composition Showcase Recital|Chamber Recital|Chamber Music Recital|Chamber Music|Organ Recital|Strings Recital|Violin & Viola|Violin/Viola|Drum Ensemble|Drum Recital|Voice Showcase Recital|Chinese Music Recital|Piano Studios Recital|Jazz Chamber|Piano Recital|Comps Fest|Recorder Recital|Music Ensemble") ~ "Studio Recital",
+    str_detect(what, "Orchestra Concert|Jazz Concert|Symphony Concert|Symphony Band|Choir Concert|Orchestra and Choir|Chinese & Global|Chinese Global Concert|Chinese and Global|Chinese Music Concert|Chinese Music Ensemble|Chinese Ensemble|Music Comps|Jazz Vocal Concert") ~ "Ensemble Concert",
+    str_detect(what, "CSA|Just Cellin|Lunar New Year|ACA|A Cappella|Accidentals|Exit 69|Date Knight|Knights|Knightingales|International Festival") ~ "Student Activity",
+    str_detect(what, "Masterclass|Lecture|Symposium") ~ "Masterclass",
+    str_detect(what, "Trustees|Trustee's|Presidents|Conference|President's|Presentation") ~ "Presentation",
+    str_detect(what, "Clinic|Music Fest|Music Department Showcase|Melinda Russell|Launch|Event|Opening") ~ "Special Events",
+      TRUE ~ "Guest"
+  )) %>% 
+  mutate(year = term_to_year(term)) %>% 
+  mutate(term = factor(term, levels = c("F15", "W16", "S16", "F16", "W17", "S17", 
+                                        "F17", "W18", "S18", "F18", "W19", "S19", 
+                                        "F19", "W20", "S20", "F20", "W21", "S21", 
+                                        "F21", "W22", "S22", "F22", "W23", "S23", 
+                                        "F23", "W24", "S24"), ordered = TRUE)) %>%
+  arrange(year, term) %>%
+  mutate(
+    term_category = case_when(
+      str_detect(term, "^F") ~ "Fall",
+      str_detect(term, "^W") ~ "Winter",
+      str_detect(term, "^S") ~ "Spring"
+    )
+  ) %>%
+  mutate(term_category = factor(term_category, levels = c("Spring", "Winter", "Fall"), ordered = TRUE))
+
+glimpse(combined_data_filtered)
+
+
+event_summary <- combined_data_filtered %>% 
+  group_by(year, term) %>% 
+  summarize(term_total = n(), .groups = 'drop') %>%
+  group_by(year) %>% 
+  mutate(year_total = sum(term_total)) %>% 
+  ungroup()
+
+```
+
+
+
+## EDA Plots
+
+# Overall Event Summary
+```{r}
+# Sort the data frame by year and term in the order of F, W, S
+event_summary <- event_summary %>%
+  mutate(term = factor(term, levels = c("F15", "W16", "S16", "F16", "W17", "S17", 
+                                        "F17", "W18", "S18", "F18", "W19", "S19", 
+                                        "F19", "W20", "S20", "F20", "W21", "S21", 
+                                        "F21", "W22", "S22", "F22", "W23", "S23", 
+                                        "F23", "W24", "S24"), ordered = TRUE)) %>%
+  arrange(year, term) %>%
+  mutate(
+    term_category = case_when(
+      str_detect(term, "^F") ~ "Fall",
+      str_detect(term, "^W") ~ "Winter",
+      str_detect(term, "^S") ~ "Spring"
+    )
+  ) %>%
+  mutate(term_category = factor(term_category, levels = c("Fall", "Winter", "Spring"), ordered = TRUE))
+
+
+# Create the stacked bar chart
+ggplot(event_summary, aes(x = year, y = term_total, fill = term_category)) +
+  geom_bar(stat = "identity") +
+  geom_text(aes(label = term_total), 
+            position = position_stack(vjust = 0.5), 
+            size = 3, 
+            color = "black") +
+  scale_fill_manual(values = c("Fall" = "#FF9999", "Winter" = "#99CCFF", "Spring" = "#99FF99")) +
+  labs(x = "Year", y = "Total Events", fill = "Term") +
+  theme_minimal() +
+  theme(axis.text.x = element_text(angle = 45, hjust = 1))
+```
+
+
+# Breakdown of Events by Support Level
+```{r}
+library(dplyr)
+library(tidyr)
+library(purrr)
+library(ggplot2)
+
+# Function to summarize and pivot data for each year
+summarize_and_pivot <- function(current_year) {
+  combined_data_filtered %>%
+    filter(year == current_year) %>%
+    group_by(term_category, support_level) %>%
+    summarize(Support = n(), .groups = 'drop') %>%
+    pivot_wider(names_from = support_level, values_from = Support, values_fill = list(Support = 0)) %>%
+    mutate(year = current_year)  # Add a year column to identify the table
+}
+
+# List of years to iterate over
+years <- unique(combined_data_filtered$year)
+
+# Use purrr::map to apply summarize_and_pivot function for each year
+support_tables <- map_dfr(years, summarize_and_pivot)  # Combine into a single data frame
+
+# Melt the data for ggplot
+support_tables_melted <- support_tables %>%
+  pivot_longer(cols = -c(term_category, year), names_to = "support_level", values_to = "Support")
+
+# Ensure the term_category and support_level are factors with the correct order
+support_tables_melted <- support_tables_melted %>%
+  filter(support_level != "NA") %>% 
+  mutate(term_category = factor(term_category, levels = c("Fall", "Winter", "Spring"), ordered = TRUE),
+         support_level = factor(support_level, levels = c("NA", "H", "M", "L"), ordered = TRUE))
+
+# Calculate percentages for each segment
+support_tables_melted <- support_tables_melted %>%
+  group_by(year, term_category) %>%
+  mutate(total_support = sum(Support),
+         percentage = (Support / total_support) * 100)
+
+# Create the facet plot
+ggplot(support_tables_melted, aes(x = term_category, y = Support, fill = support_level)) +
+  geom_bar(stat = "identity", position = "stack") +
+  geom_text(aes(label = paste0(Support, " (", round(percentage, 1), "%)")),
+            position = position_stack(vjust = 0.5), size = 2, color = "black") +
+  facet_wrap(~ year) +
+  labs(x = "Term", y = "Support Count", fill = "Support Level") +
+  scale_fill_manual(values = c("H" = "#FF9999", "M" = "#99CCFF", "L" = "#99FF99", "NA" = "black")) +
+  theme_minimal() +
+  theme(axis.text.x = element_text(angle = 45, hjust = 1))
+
+```
+
+
+# Breakdown of Events by Department/Source
+
+```{r}
+# Function to summarize and pivot data for each year
+summarize_and_pivot_department <- function(current_year) {
+  combined_data_filtered %>%
+    filter(year == current_year) %>%
+    group_by(term_category, department_type) %>%
+    summarize(DepartmentCount = n(), .groups = 'drop') %>%
+    pivot_wider(names_from = department_type, values_from = DepartmentCount, values_fill = list(DepartmentCount = 0)) %>%
+    mutate(year = current_year)  # Add a year column to identify the table
+}
+
+# List of years to iterate over
+years <- unique(combined_data_filtered$year)
+
+# Use purrr::map to apply summarize_and_pivot function for each year
+department_tables <- map_dfr(years, summarize_and_pivot_department)  # Combine into a single data frame
+
+# Melt the data for ggplot
+department_tables_melted <- department_tables %>%
+  pivot_longer(cols = -c(term_category, year), names_to = "department_type", values_to = "DepartmentCount")
+
+# Ensure the term_category and department_type are factors with the correct order
+department_tables_melted <- department_tables_melted %>%
+  mutate(term_category = factor(term_category, levels = c("Fall", "Winter", "Spring"), ordered = TRUE),
+         department_type = factor(department_type, levels = c("MUSC", "ODOA", "CSA", "Collab", "Others"), ordered = TRUE))
+
+# Filter out NA department types (if any)
+department_tables_melted <- department_tables_melted %>%
+  filter(department_type != "NA")
+
+# Calculate percentages for each segment
+department_tables_melted <- department_tables_melted %>%
+  group_by(year, term_category) %>%
+  mutate(total_count = sum(DepartmentCount, na.rm = T),
+         percentage = (DepartmentCount / total_count) * 100)
+
+# Create the facet plot
+ggplot(department_tables_melted, aes(x = term_category, y = DepartmentCount, fill = department_type)) +
+  geom_bar(stat = "identity", position = "stack") +
+  geom_text(aes(label = paste0(DepartmentCount, " (", round(percentage, 1), "%)")),
+            position = position_stack(vjust = 0.5), size = 2, color = "black",
+            check_overlap = TRUE) +
+  facet_wrap(~ year) +
+  labs(x = "Term", y = "Department Count", fill = "Department Type") +
+  scale_fill_manual(values = c("MUSC" = "#FF9999", "ODOA" = "#99CCFF", "CSA" = "#99FF99", "Collab" = "#FFD700", "Others" = "#FFA500")) +
+  theme_minimal() +
+  theme(axis.text.x = element_text(angle = 45, hjust = 1))
+
+```
+
+
+# Breakdown of Music, Collab & ODOA Events by Type
+```{r}
+# Function to summarize and pivot data for each year
+summarize_and_pivot_event <- function(current_year) {
+  combined_data_filtered %>%
+    filter(year == current_year) %>%
+    group_by(term_category, event_type) %>%
+    summarize(EventCount = n(), .groups = 'drop') %>%
+    pivot_wider(names_from = event_type, values_from = EventCount, values_fill = list(EventCount = 0)) %>%
+    mutate(year = current_year)  # Add a year column to identify the table
+}
+
+# List of years to iterate over
+years <- unique(combined_data_filtered$year)
+
+# Use purrr::map to apply summarize_and_pivot function for each year
+event_tables <- map_dfr(years, summarize_and_pivot_event)  # Combine into a single data frame
+
+# Melt the data for ggplot
+event_tables_melted <- event_tables %>%
+  pivot_longer(cols = -c(term_category, year), names_to = "event_type", values_to = "EventCount")
+
+# Ensure the term_category and event_type are factors with the correct order
+event_tables_melted <- event_tables_melted %>%
+  mutate(term_category = factor(term_category, levels = c("Fall", "Winter", "Spring"), ordered = TRUE),
+         event_type = factor(event_type, levels = c("Ensemble Concert", "Student Activity", "Studio Recital",
+                                                    "Guest", "Faculty Recital", "Student Recital",
+                                                    "Special Events", "Presentation", "Masterclass"),
+                            ordered = TRUE))
+
+# Calculate percentages for each segment
+event_tables_melted <- event_tables_melted %>%
+  group_by(year, term_category) %>%
+  mutate(total_count = sum(EventCount, na.rm = T),
+         percentage = (EventCount / total_count) * 100)
+
+# Create the facet plot
+ggplot(event_tables_melted, aes(x = term_category, y = EventCount, fill = event_type)) +
+  geom_bar(stat = "identity", position = "stack") +
+  geom_text(aes(label = paste0(EventCount, " (", round(percentage, 1), "%)")),
+            position = position_stack(vjust = 0.5), size = 2, color = "black",
+            check_overlap = TRUE) +
+  facet_wrap(~ year) +
+  labs(x = "Term", y = "Event Count", fill = "Event Type") +
+  scale_fill_brewer(palette = "Set2") +  # Using a Brewer palette for default colors
+  theme_minimal() +
+  theme(axis.text.x = element_text(angle = 45, hjust = 1))
+
+```
+

Event Data Analysis.Rproj

@@ -0,0 +1,13 @@
+Version: 1.0
+
+RestoreWorkspace: Default
+SaveWorkspace: Default
+AlwaysSaveHistory: Default
+
+EnableCodeIndexing: Yes
+UseSpacesForTab: Yes
+NumSpacesForTab: 2
+Encoding: UTF-8
+
+RnwWeave: Sweave
+LaTeX: pdfLaTeX