written narrative

By Tenzin Dhondup, Nimo Hassan, Breanna Ranglall, Alexander Wodarski

For our final project in DASC 130, we partnered with the Office of Sustainability Initiatives (OSI) to explore key sustainability credit areas. Our group focused on OP-18: Waste Minimization and Diversion, which evaluates waste reduction efforts in higher education institutions. We chose this area because we were interested in seeing how the University of St. Thomas performs in waste management compared to other Minnesota peer institutions, especially given the university’s investments in sustainability initiatives.

To conduct our analysis, we used the AASHE STARS platform—a comprehensive tool that provides transparent and standardized metrics for assessing sustainability performance across colleges and universities. Through this platform, we examined changes in waste minimization over time and evaluated how St. Thomas measures up against its peers. Our findings are presented in a visually engaging and reproducible format, offering actionable insights into sustainability efforts at St. Thomas and beyond.

Data Tidying/Wrangling

# Loading needed libraries 
library(tidyverse)
library(readxl)
library(dplyr)

Read data from excel spreadsheets

# Read data from excel spreadsheet

dataf <- read_excel("OP-18_Waste_Minimization_and_Diversion_2024.xlsx")
pop_data <- read.csv("updated_school_population.csv")

Converting ‘chr’ to ‘numeric’

dataframe <- dataf

# Converting 'chr' to 'numeric'

dataframe$'Materials recycled, performance year' <- as.numeric(dataframe$'Materials recycled, performance year')
dataframe$'Materials recycled, baseline year' <- as.numeric(dataframe$'Materials recycled, baseline year')
dataframe$'Materials composted, performance year' <- as.numeric(dataframe$'Materials composted, performance year')
dataframe$'Materials composted, baseline year' <- as.numeric(dataframe$'Materials composted, baseline year')
dataframe$'Materials donated or re-sold, performance year' <- as.numeric(dataframe$'Materials donated or re-sold, performance year')
dataframe$'Materials donated or re-sold, baseline year' <- as.numeric(dataframe$'Materials donated or re-sold, baseline year')
dataframe$'Materials disposed through post-recycling residual conversion, performance year' <- as.numeric(dataframe$'Materials disposed through post-recycling residual conversion, performance year')
dataframe$'Materials disposed through post-recycling residual conversion, baseline year' <- as.numeric(dataframe$'Materials disposed through post-recycling residual conversion, baseline year')
dataframe$'Materials disposed in a solid waste landfill or incinerator, performance year' <- as.numeric(dataframe$'Materials disposed in a solid waste landfill or incinerator, performance year')
dataframe$'Materials disposed in a solid waste landfill or incinerator, baseline year' <- as.numeric(dataframe$'Materials disposed in a solid waste landfill or incinerator, baseline year')
dataframe$'Total waste generated, performance year' <- as.numeric(dataframe$'Total waste generated, performance year')
dataframe$'Total waste generated, baseline year' <- as.numeric(dataframe$'Total waste generated, baseline year')
dataframe$'Start date, performance year or 3-year period' <- as.numeric(dataframe$'Start date, performance year or 3-year period')
dataframe$'End date, performance year or 3-year period' <- as.numeric(dataframe$'End date, performance year or 3-year period')
dataframe$'Start date, baseline year or 3-year period' <- as.numeric(dataframe$'Start date, baseline year or 3-year period')
dataframe$'End date, baseline year or 3-year period' <- as.numeric(dataframe$'End date, baseline year or 3-year period')
dataframe$'Number of students resident on-site, performance year' <- as.numeric(dataframe$'Number of students resident on-site, performance year')
dataframe$'Number of students resident on-site, baseline year' <- as.numeric(dataframe$'Number of students resident on-site, baseline year')
dataframe$'Number of employees resident on-site, performance year' <- as.numeric(dataframe$'Number of employees resident on-site, performance year')
dataframe$'Number of employees resident on-site, baseline year' <- as.numeric(dataframe$'Number of employees resident on-site, baseline year')
dataframe$'Number of other individuals resident on-site, performance year' <- as.numeric(dataframe$'Number of other individuals resident on-site, performance year')
dataframe$'Number of other individuals resident on-site, baseline year' <- as.numeric(dataframe$'Number of other individuals resident on-site, baseline year')
dataframe$'Total full-time equivalent student enrollment, performance year' <- as.numeric(dataframe$'Total full-time equivalent student enrollment, performance year')
dataframe$'Total full-time equivalent student enrollment, baseline year' <- as.numeric(dataframe$'Total full-time equivalent student enrollment, baseline year')
dataframe$'Full-time equivalent of employees, performance year' <- as.numeric(dataframe$'Full-time equivalent of employees, performance year')
dataframe$'Full-time equivalent of employees, baseline year' <- as.numeric(dataframe$'Full-time equivalent of employees, baseline year')
dataframe$'Full-time equivalent of students enrolled in exclusively in distance education, performance year' <- as.numeric(dataframe$'Full-time equivalent of students enrolled in exclusively in distance education, performance year')
dataframe$'Full-time equivalent of students enrolled exclusively in distance education, baseline year' <- as.numeric(dataframe$'Full-time equivalent of students enrolled exclusively in distance education, baseline year')
dataframe$'Weighted campus users, performance year' <- as.numeric(dataframe$'Weighted campus users, performance year')
dataframe$'Weighted campus users, baseline year' <- as.numeric(dataframe$'Weighted campus users, baseline year')
dataframe$'Total waste generated per weighted campus user, performance year' <- as.numeric(dataframe$'Total waste generated per weighted campus user, performance year')
dataframe$'Total waste generated per weighted campus user, baseline year' <- as.numeric(dataframe$'Total waste generated per weighted campus user, baseline year')
dataframe$'Weighted campus users, performance year' <- as.numeric(dataframe$'Weighted campus users, performance year')
dataframe$'Weighted campus users, baseline year' <- as.numeric(dataframe$'Weighted campus users, baseline year')
dataframe$'Percentage reduction in total waste generated per weighted campus user from baseline' <- as.numeric(dataframe$'Percentage reduction in total waste generated per weighted campus user from baseline')
dataframe$'Percentage of materials diverted from the landfill or incinerator by recycling, composting, ...' <- as.numeric(dataframe$'Percentage of materials diverted from the landfill or incinerator by recycling, composting, ...')
dataframe$'Percentage of materials diverted from the landfill or incinerator (including up to 10 percent ...' <- as.numeric(dataframe$'Percentage of materials diverted from the landfill or incinerator (including up to 10 percent ...')
# Replace the single NA with 0
# Replace NA values with 0 for the specified columns
dataframe$`Total waste generated, baseline year`[is.na(dataframe$`Total waste generated, baseline year`)] <- 0
dataframe$`Weighted campus users, baseline year`[is.na(dataframe$`Weighted campus users, baseline year`)] <- 0
dataframe$`Total waste generated, performance year`[is.na(dataframe$`Total waste generated, performance year`)] <- 0
dataframe$`Weighted campus users, performance year`[is.na(dataframe$`Weighted campus users, performance year`)] <- 0
dataframe$`Materials recycled, performance year`[is.na(dataframe$`Materials recycled, performance year`)] <- 0
dataframe$`Materials composted, performance year`[is.na(dataframe$`Materials composted, performance year`)] <- 0
dataframe$`Materials donated or re-sold, performance year`[is.na(dataframe$`Materials donated or re-sold, performance year`)] <- 0
dataframe$`Materials disposed in a solid waste landfill or incinerator, performance year`[is.na(dataframe$`Materials disposed in a solid waste landfill or incinerator, performance year`)] <- 0

Rename some Universities

pop_data$Institution[pop_data$Institution == "University of Minnesota Twin Cities"] <- "University of Minnesota, Twin Cities"
pop_data$Institution[pop_data$Institution == "University of Minnesota Crookston"] <- "University of Minnesota, Crookston"
pop_data$Institution[pop_data$Institution == "University of Minnesota Morris"] <- "University of Minnesota, Morris"

Converting ‘chr’ to ‘date’

dataframe$'Last Updated' <- as.Date(dataframe$'Last Updated')
dataframe$'Start date, performance year or 3-year period' <- as.Date(dataframe$'Start date, performance year or 3-year period', origin = "1899-12-30")
dataframe$'End date, performance year or 3-year period' <- as.Date(dataframe$'End date, performance year or 3-year period', origin = "1899-12-30")
dataframe$'Start date, baseline year or 3-year period' <- as.Date(dataframe$'Start date, baseline year or 3-year period', origin = "1899-12-30")
dataframe$'End date, baseline year or 3-year period' <- as.Date(dataframe$'End date, baseline year or 3-year period', origin = "1899-12-30")
pop_data$Undergraduate.Population <- as.numeric(pop_data$Undergraduate.Population)

Combine Both Spreadsheets

peer_schools <- c("Iowa State University", "University of Minnesota, Twin Cities", "University of St. Thomas", 
                  "Macalester College", "Gustavus Adolphus College", "University of Wisconsin-Madison", 
                  "Marquette University", "University of Wisconsin-Eau Claire", 
                  "University of Wisconsin-Green Bay", "University of Wisconsin-Oshkosh", 
                  "University of Wisconsin-Platteville", "University of Wisconsin-Stevens Point", 
                  "University of Wisconsin-River Falls", "University of Minnesota, Morris", 
                  "University of Minnesota, Crookston","Concordia University",
                  "Carleton College", "South Dakota State University", 
                  "University of South Dakota")
schools_to_exclude <- c("Marquette University", "University of Wisconsin-Green Bay", "University of Minnesota, Crookston")
peers <-  dataframe|>
  filter(Institution %in% peer_schools)|>
  left_join(pop_data,by = "Institution")|>
  filter(Undergraduate.Population<11000)|>
  filter(!Institution %in% schools_to_exclude)|>
  mutate(
  `Reduction in Total Waste per Person` = 5 * (
    (
      (`Total waste generated, baseline year` / `Weighted campus users, baseline year`) - 
      (`Total waste generated, performance year` / `Weighted campus users, performance year`)
    ) / 
    (`Total waste generated, baseline year` / `Weighted campus users, baseline year`)
  ),
  
  `Total waste per person` = 2.78 * (
    (
      0.46 - 
      (`Total waste generated, performance year` / `Weighted campus users, performance year`) / 0.46
    )
  ),
  
  `Waste diverted from the landfill or incinerator` = 3 * (
    (
      (
        `Materials recycled, performance year` + 
        `Materials composted, performance year` + 
        `Materials donated or re-sold, performance year` +
        (0.1 * (
          `Materials recycled, performance year` + 
          `Materials composted, performance year` + 
          `Materials donated or re-sold, performance year` + 
          `Materials disposed in a solid waste landfill or incinerator, performance year`
        ))
      ) / 
      (
        `Materials recycled, performance year` + 
        `Materials composted, performance year` + 
        `Materials donated or re-sold, performance year` + 
        `Materials disposed in a solid waste landfill or incinerator, performance year`
      )
    )
  )
)

Summary Statistics

While analyzing Minnesota peer institutions, we noticed that there weren’t many schools to compare. To expand our analysis, we included schools from North Dakota, South Dakota, Iowa, and Wisconsin, creating a broader dataset. We compiled a new CSV file with these schools, including their undergraduate populations, and filtered the data accordingly. To address outliers, we excluded any schools with undergraduate populations over 11,000. Additionally, we filtered out schools that did not report data or lacked sufficient information to calculate their scores, placing them on an exclusion list. Finally, we calculated the scores using equations from the AASHE STARS website. This required some adjustments, such as replacing NA values with 0, to ensure that each institution generated a score.

When looking at the summary statistics, an important thing to note is that all variables are measured in tons. We see that on average the schools in Iowa, Minnesota, South Dakota, and Wisconsin, within our undergraduate population range, dispose of 587 tons of solid waste in landfills, compared to just recycling 278 tons of waste. With that number being about double, these schools could improve their efforts to recycle more, in order to be a more sustainable institution.

numeric_summary <- peers |>
  summarize(across(where(is.numeric), list(
    mean = ~ mean(.x, na.rm = TRUE),
    median = ~ median(.x, na.rm = TRUE),
    sd = ~ sd(.x, na.rm = TRUE),
    min = ~ min(.x, na.rm = TRUE),
    max = ~ max(.x, na.rm = TRUE)
  )))
View(peers)

library(dplyr)
library(tidyr)


summarydata <- data.frame(
  Material = rep(c(
    "Materials recycled",
    "Materials composted",
    "Materials donated or re-sold",
    "Materials disposed through post-recycling residual conversion",
    "Materials disposed in a solid waste landfill or incinerator"
  ), each = 10),
  Year_Type = rep(c(
    "Performance Year", "Baseline Year"
  ), times = 5),
  Metric = rep(c("mean", "median", "sd", "min", "max"), times = 10),
  Value = c(
    2.2, 2.2, 2.2, 0, 2.2,
    278.7805, 256.3863, 183.5492, 41.03, 773.5,
    307.247, 318, 172.3574, 22.45, 531.64,
    77.07808, 67.4, 58.18791, 0, 188,
    95.93739, 67.55, 77.5185, 7, 185.6,
    34.03183, 6.6, 50.33896, 0, 134.7,
    105.936, 73.55, 87.61951, 35.6, 257,
    NA, NA, NA, Inf, -Inf,
    NA, NA, NA, Inf, -Inf,
    587.6389, 422, 417.0653, 104, 1534.84
  )
)


organizedSummary <- summarydata|>
  pivot_wider(
    names_from = Metric,
    values_from = Value
  ) |>
  arrange(Material, Year_Type)


view(organizedSummary)

Data Visuilzations

The next step in our analysis was to interpret the data to understand trends and comparisons better visually. Data visualizations allowed us to uncover patterns, identify outliers, and draw meaningful insights that may not be immediately evident from raw numbers alone. The Key findings are shown in the following visualizations using charts and graphs, which also show the overall patterns found in the dataset and how St. Thomas measures up against other institutions in terms of waste minimization and diversion initiatives. These graphics show the university’s performance more clearly and point out areas that could use improvement.

The analysis examines the relationship between total waste generated and the full-time equivalent (FTE) student enrollment for a selection of institutions during the performance year. The dataset was filtered to include institutions of interest, such as the University of St. Thomas, Carleton College, Gustavus Adolphus College, Macalester College, and various universities from South Dakota and Wisconsin. Only data for “Total waste generated” and “Total full-time equivalent student enrollment” were considered, and rows with missing values were excluded. The two datasets were then merged using institution names as the key, resulting in a combined dataset containing waste generation and enrollment figures for each institution.

A scatterplot was created to visualize the relationship between FTE enrollment and waste generation, with each point representing an institution. The x-axis displays FTE student enrollment, while the y-axis shows the total waste generated during the same period. Institutions are distinguished by color, making it easier to identify trends or outliers. The visualization reveals how student population size correlates with waste generation. While larger institutions are generally expected to produce more waste, deviations from this pattern could indicate differences in waste management efficiency. Institutions with higher waste levels relative to enrollment may have opportunities to improve sustainability practices, whereas those with lower waste levels may already be implementing effective measures.

 # Converting df from wide format to long format

df_long <- dataframe |>
  select(Institution, where(is.numeric)) |>
  pivot_longer(
    cols = starts_with("Materials") | starts_with("Total") | starts_with("Number") | starts_with("Full-Time") | starts_with("Weighted"),
    names_to = c("Variable_Type", "Year"),
    names_sep = ", ",
    values_to = "Amount"
  )
  
  #Combining Tidy Numeric Data with Non Numeric Data

df_non_numeric <- dataframe |>
  select(where(~ !is.numeric(.)))

data_test <- left_join(df_long, df_non_numeric, by = "Institution")

# Filter waste and enrollment data separately, then join them
waste_data <- data_test |>
  filter(Institution %in% c("University of St. Thomas", "Carleton College", "Concordia University", "   
Gustavus Adolphus College", "Macalester College", "South Dakota State University", "University of South Dakota", "University of Wisconsin-Eau Claire", "University of Wisconsin-Oshkosh", " 
University of Wisconsin-Platteville", "University of Wisconsin-River Falls", "University of Wisconsin-Stevens Point")) |>
  filter(Variable_Type == "Total waste generated") |>
  filter(Year == "performance year") |>
  filter(!is.na(Amount)) |>
  select(Institution, Amount, Variable_Type, Year)

enrollment_data <- data_test |>
  filter(Institution %in% c("University of St. Thomas", "Carleton College", "   
Gustavus Adolphus College", "Macalester College", "South Dakota State University", "University of South Dakota", "University of Wisconsin-Eau Claire", "University of Wisconsin-Oshkosh", " 
University of Wisconsin-Platteville", "University of Wisconsin-River Falls", "University of Wisconsin-Stevens Point")) |>
  filter(Variable_Type == "Total full-time equivalent student enrollment") |>
  filter(Year == "performance year") |>
  filter(!is.na(Amount)) |>
  select(Institution, Amount, Variable_Type, Year)

# Join the two datasets on Institution
merged_data <- left_join(waste_data, enrollment_data, by = "Institution", suffix = c("_waste", "_enrollment"))

# Plot the data with a legend
ggplot(merged_data, aes(x = Amount_enrollment, y = Amount_waste)) +
  geom_point(aes(color = Institution), size = 3) +  # Mapping color to Institution for legend
  labs(title = "Waste Generation vs. Campus Population",
       x = "Full-Time Equivalent Students (Performance Year)",
       y = "Total Waste Generated (Performance Year)",
       color = "Institution",
       caption = "Comparison of Total Waste Generation and Campus Population Across Institutions: \nEach point represents an institution, \nwith its total waste generated during the performance year \nplotted against the full-time equivalent student population for the same period.") +  # Legend title
  theme_minimal()

The analysis focuses on visualizing the total waste generated by various institutions during the performance year using a bar chart. The dataset includes key institutions such as the University of St. Thomas, Carleton College, Gustavus Adolphus College, Macalester College, and several universities from South Dakota and Wisconsin. The bar chart represents the total waste generated by each institution, with the University of St. Thomas specifically highlighted in purple for emphasis. This visual distinction draws attention to the University of St. Thomas while allowing for easy comparison with other institutions.

Each bar corresponds to the waste generated by an institution, with the y-axis representing the total waste amount and the x-axis displaying the institution names. The chart effectively highlights differences in waste production across campuses, making it clear which institutions generate more or less waste relative to others. By angling the institution names on the x-axis, the chart improves readability, particularly when comparing multiple institutions.

# Bar Plot: Total Waste Generated by Institution
ggplot(merged_data, aes(x = Institution, y = Amount_waste, fill = Institution)) +
  geom_bar(stat = "identity", show.legend = FALSE) +
  scale_fill_manual(values = c("University of St. Thomas" = "purple"))+
  labs(title = "Total Waste Generated by Institution",
       x = "Institution",
       y = "Total Waste Generated (Performance Year)",
       caption = "Total Waste Generated by Institution: \nThis bar chart shows the total waste generated by each institution during the performance year, \nhighlighting variations in waste production across campuses. \nThe University of St. Thomas is emphasized in purple for specific focus.") +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

The University of St. Thomas excels in recycling and landfill diversion compared to the overall average of other schools. However, it composts and donates/resells less, highlighting areas for potential improvement in waste management practices.

# Filter relevant columns for visualization
filtered_data_viz3 <- peers |>
  select(
    Institution,
    `Materials recycled, performance year`,
    `Materials composted, performance year`,
    `Materials donated or re-sold, performance year`,
    `Materials disposed in a solid waste landfill or incinerator, performance year`
  )

# Rename columns for simplicity
colnames(filtered_data_viz3) <- c(
  'Institution', 
  'Recycled', 
  'Composted', 
  'Donated or Resold', 
  'Disposed'
)

# pivot the dataframe to long format for visualization
pivot_data <- filtered_data_viz3 |>
  pivot_longer(
    cols = c('Recycled', 'Composted', 'Donated or Resold', 'Disposed'),
    names_to = 'Material_Type',
    values_to = 'Amount'
  )

ggplot(pivot_data, aes(x = Institution, y = Amount, fill = Material_Type)) +
  geom_bar(stat = 'identity') +
  scale_fill_manual(
    values = c(
      'Recycled' = 'blue',
      'Composted' = 'darkgreen',
      'Donated or Resold' = 'orange',
      'Disposed' = 'red'
    )
  ) +
  labs(
    x = 'Institution',
    y = 'Total Amount of Material (tons)',
    title = 'Material Disposal Methods by Institution',
    fill = 'Material Type'
  ) +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 90, hjust = 1))

The University of St. Thomas demonstrates relatively higher waste reduction compared to other institutions, with its population size aligning closely with the majority. This indicates effective waste reduction efforts relative to its student population.

filtered_data_viz4 <- peers|>
  select(Institution, Undergraduate.Population, 'Percentage reduction in total waste generated per weighted campus user from baseline') |>
  rename(
    School = Institution,
    Population = Undergraduate.Population,
    Total_Waste_Reduction = 'Percentage reduction in total waste generated per weighted campus user from baseline'
  ) |>
  filter(!is.na(School) & !is.na(Population) & !is.na(Total_Waste_Reduction))

# Create a column to distinguish University of St. Thomas from other schools
filtered_data_viz4 <- filtered_data_viz4 |>
  mutate(Color = ifelse(grepl('University of St. Thomas', School), 'purple', 'grey'))

# Create the scatter plot
ggplot(filtered_data_viz4, aes(x = School, y = Total_Waste_Reduction, size = Population, color = Color)) +
  geom_point(shape = 21, fill = filtered_data_viz4$Color) +  # Change shape to circle (shape = 21) or square (shape = 22)
  scale_color_manual(values = c('purple' = 'purple', 'grey' = 'grey')) +
  scale_size(range = c(3, 10)) +  # Adjust the size of the points
  labs(
    x = 'School',
    y = 'Total Reduction in Waste',
    title = 'Comparison of Schools by Total Waste Reduction Percentage'
  ) +
  theme_minimal() +
  theme(
    axis.text.x = element_text(angle = 90, hjust = 1),
    legend.position="none"
  )