# List of packages
 packages <- c("tidyverse", "gt", "gapminder", "srvyr", "fst", "ggridges")
 
 # Install packages if they aren't installed already
new_packages <- packages[!(packages %in% installed.packages()[,"Package"])]
 if(length(new_packages)) install.packages(new_packages)
 
 # Load the packages
 lapply(packages, library, character.only = TRUE)
## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr     1.1.4     ✔ readr     2.1.5
## ✔ forcats   1.0.0     ✔ stringr   1.5.1
## ✔ ggplot2   3.5.1     ✔ tibble    3.2.1
## ✔ lubridate 1.9.4     ✔ tidyr     1.3.1
## ✔ purrr     1.0.2     
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag()    masks stats::lag()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors
## 
## Attaching package: 'srvyr'
## 
## 
## The following object is masked from 'package:stats':
## 
##     filter
## [[1]]
##  [1] "lubridate" "forcats"   "stringr"   "dplyr"     "purrr"     "readr"    
##  [7] "tidyr"     "tibble"    "ggplot2"   "tidyverse" "stats"     "graphics" 
## [13] "grDevices" "utils"     "datasets"  "methods"   "base"     
## 
## [[2]]
##  [1] "gt"        "lubridate" "forcats"   "stringr"   "dplyr"     "purrr"    
##  [7] "readr"     "tidyr"     "tibble"    "ggplot2"   "tidyverse" "stats"    
## [13] "graphics"  "grDevices" "utils"     "datasets"  "methods"   "base"     
## 
## [[3]]
##  [1] "gapminder" "gt"        "lubridate" "forcats"   "stringr"   "dplyr"    
##  [7] "purrr"     "readr"     "tidyr"     "tibble"    "ggplot2"   "tidyverse"
## [13] "stats"     "graphics"  "grDevices" "utils"     "datasets"  "methods"  
## [19] "base"     
## 
## [[4]]
##  [1] "srvyr"     "gapminder" "gt"        "lubridate" "forcats"   "stringr"  
##  [7] "dplyr"     "purrr"     "readr"     "tidyr"     "tibble"    "ggplot2"  
## [13] "tidyverse" "stats"     "graphics"  "grDevices" "utils"     "datasets" 
## [19] "methods"   "base"     
## 
## [[5]]
##  [1] "fst"       "srvyr"     "gapminder" "gt"        "lubridate" "forcats"  
##  [7] "stringr"   "dplyr"     "purrr"     "readr"     "tidyr"     "tibble"   
## [13] "ggplot2"   "tidyverse" "stats"     "graphics"  "grDevices" "utils"    
## [19] "datasets"  "methods"   "base"     
## 
## [[6]]
##  [1] "ggridges"  "fst"       "srvyr"     "gapminder" "gt"        "lubridate"
##  [7] "forcats"   "stringr"   "dplyr"     "purrr"     "readr"     "tidyr"    
## [13] "tibble"    "ggplot2"   "tidyverse" "stats"     "graphics"  "grDevices"
## [19] "utils"     "datasets"  "methods"   "base"
  1. Data Manipulation (gapminder data) • Filter for years 1987 and 2007 • Calculate mean life expectancy by continent for each year • Calculate the change between years • In a separate step, filter to five focal countries (Niger, Bangladesh, El Salvador, Iraq, Zimbabwe)
# Filter data for years 1987 and 2007 
gap_filtered <- gapminder %>%
  filter(year %in% c(1987, 2007))
# Calculate mean life expectancy by continent for each year
 mean_lifeExp <- gap_filtered %>%
  group_by(continent, year) %>%
  summarise(mean_lifeExp = mean(lifeExp), .groups = "drop")
 # Reshape data and calculate the change in life expectancy
 lifeExp_change <- mean_lifeExp %>%
  spread(year, mean_lifeExp) %>%
  mutate(change = `2007` - `1987`)
 
 # print the results
 print(lifeExp_change)
## # A tibble: 5 × 4
##   continent `1987` `2007` change
##   <fct>      <dbl>  <dbl>  <dbl>
## 1 Africa      53.3   54.8   1.46
## 2 Americas    68.1   73.6   5.52
## 3 Asia        64.9   70.7   5.88
## 4 Europe      73.6   77.6   4.01
## 5 Oceania     75.3   80.7   5.40

lines 33-35 groups the data by continent and year and calculates the average life expectancy (mean(lifeExp)) for each continent in both years. spread(year, mean_lifeExp) converts the year column into seperate columns (1987 and 2007), with the mean life expectancy as values. mutate(change= 2007-1987) creates a new column “change” that calculates the difference in life expectancy between 2007 and 1987.

# Filter the lifeExp_change table for the five focal countries
focal_countries_table <- gap_filtered %>%
  filter(country %in% c("Niger", "Bangladesh", "El Salvador", "Iraq", "Zimbabwe")) %>%
  group_by(country, year) %>%
  summarise(mean_lifeExp = mean(lifeExp), .groups = "drop") %>%
  spread(year, mean_lifeExp) %>%
  mutate(Change = `2007` - `1987`)

# Display the filtered table for the focal countries
focal_countries_table
## # A tibble: 5 × 4
##   country     `1987` `2007` Change
##   <fct>        <dbl>  <dbl>  <dbl>
## 1 Bangladesh    52.8   64.1  11.2 
## 2 El Salvador   63.2   71.9   8.72
## 3 Iraq          65.0   59.5  -5.50
## 4 Niger         44.6   56.9  12.3 
## 5 Zimbabwe      62.4   43.5 -18.9

The dataset gap_filtered is filtered to keep only rows where the country column matches one of the five focal countries. This data is then grouped by country and year. Within each country-year group, the mean of the lifeExp column is calculated. The .group=“drop” ensures that after summarization, the grouping is removed, meaning the resulting dataset is no longer grouped. The wide format shows each column represents a year and contains the corresponding mean life expectancy for that country.

  1. Table Creation: • Title: Life Expectancy Changes by Continent • Subtitle: Average life expectancy in years • Columns: o 1987 values o 2007 values o Change (2007-1987) Format: o All values to one decimal place o Bold headers (Continent, 1987, 2007, Change) o Order continents by magnitude of change (largest to smallest)
formatted_table <- lifeExp_change %>%
  select(continent, `1987`, `2007`, change) %>%  # Ensure correct columns
  gt() %>%
  tab_header(
    title = md("**Life Expectancy Changes by Continent**"),  
    subtitle = md("*Average life expectancy in years*")  
  ) %>%
  cols_label(
    continent = "Continent",
    `1987` = "1987 values",
    `2007` = "2007 values",
    change = "Change (2007-1987)"
  ) %>%
  fmt_number(
    columns = c(`1987`, `2007`, change),  
    decimals = 1
  ) %>%
  tab_style(
    style = cell_text(weight = "bold"),
    locations = cells_column_labels(everything())  
  )

# Display the formatted table
formatted_table
Life Expectancy Changes by Continent
Average life expectancy in years
Continent 1987 values 2007 values Change (2007-1987)
Africa 53.3 54.8 1.5
Americas 68.1 73.6 5.5
Asia 64.9 70.7 5.9
Europe 73.6 77.6 4.0
Oceania 75.3 80.7 5.4

cols_label will list what I want each column to be named and if I want to change that name. decimal= includes what I want the decial point to.

  1. Data Visualization: • Title: “Life Expectancy Trajectories (1987-2007)” • Subtitle: “in Selected Countries” • Format: o Lines with size = 1.5 o Use scale_color_brewer(palette = “Set1”) o theme_minimal() with no minor grid lines o Legend at bottom o Title in bold (size = 14) o Subtitle size = 12 o Clear axis labels (“Year”, “Life Expectancy (years)”)
# Filter the data for the five selected countries (Niger, Bangladesh, El Salvador, Iraq, Zimbabwe)
focal_countries <- gap_filtered %>%
  filter(country %in% c("Niger", "Bangladesh", "El Salvador", "Iraq", "Zimbabwe"))

 # Create the plot using ggplot2
lifeExp_plot <- ggplot(focal_countries, aes(x = year, y = lifeExp, color = country, group = country)) +
  geom_line(size = 1.5) +  # Line thickness
  scale_color_brewer(palette = "Set1") +  # Use color palette from RColorBrewer
  theme_minimal() +  # Use minimal theme (no background grid)
  theme(
    panel.grid.minor = element_blank(),  # Remove minor grid lines
    legend.position = "bottom",  # Position the legend at the bottom
    plot.title = element_text(face = "bold", size = 14),  # Make the title bold and set size
    plot.subtitle = element_text(size = 12)  # Set subtitle size
  ) +
  labs(
    title = "Life Expectancy Trajectories (1987-2007)",  # Title of the plot
    subtitle = "in Selected Countries",  # Subtitle
    x = "Year",  # X-axis label
    y = "Life Expectancy (years)",  # Y-axis label
    color = "Country"  # Legend label for the color
  )
## Warning: Using `size` aesthetic for lines was deprecated in ggplot2 3.4.0.
## ℹ Please use `linewidth` instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.
# 3. Display the plot
print(lifeExp_plot)

This chunk is creating a line plot using ggplot2 to visualize changes in life expectancy over time (1987-2007) for five selected countries. gap_filtered selects only 5 coutries ggplot2 to plot life exxpectancy over time the colour palette chosen is RColorBrewer

I decided to group all of this information into the same chunk, as opposed to multiple chunks, to make sure it ran good while all together and provided the desired look of the graph.

  1. Interpretation:
  1. Continental Trends
  1. Five-Country Analysis

Task 2: Interpersonal Trust Patterns (3 points) a. Data Manipulation (anes_2020 data) • Remove missing values for the variables TrustPeople and AgeGroup • Calculate percentage of trust categories by age group

load("anes_2020.rda")
table(anes_2020$TrustPeople)
## 
##              Always    Most of the time About half the time    Some of the time 
##                  48                3511                2020                1597 
##               Never 
##                 264

I am making sure I load the anes data set from my seperate folder.

# Filter out rows where 'TrustPeople' or 'AgeGroup' are missing
anes_clean <- anes_2020 %>%
  filter(!is.na(TrustPeople) & !is.na(AgeGroup))  # Remove rows where either TrustPeople or AgeGroup is NA

# Store the total valid responses (the total number of responses for each AgeGroup)
total_valid_by_age <- anes_clean %>%
  group_by(AgeGroup) %>%
  summarise(total_valid = n())

# Now, calculate counts and percentages by AgeGroup and TrustPeople
trust_by_age <- anes_clean %>%
  group_by(AgeGroup, TrustPeople) %>%  # Group by both AgeGroup and TrustPeople
  summarise(count = n(), .groups = "drop") %>%  # Count responses in each group
  left_join(total_valid_by_age, by = "AgeGroup") %>%  # Join with the total valid responses by AgeGroup
  mutate(percentage = round(100 * count / total_valid, 1))  # Calculate percentage for each category within AgeGroup

# Print the result to check
print(trust_by_age)
## # A tibble: 30 × 5
##    AgeGroup TrustPeople         count total_valid percentage
##    <fct>    <fct>               <int>       <int>      <dbl>
##  1 18-29    Always                  7         871        0.8
##  2 18-29    Most of the time      268         871       30.8
##  3 18-29    About half the time   278         871       31.9
##  4 18-29    Some of the time      246         871       28.2
##  5 18-29    Never                  72         871        8.3
##  6 30-39    Always                 10        1239        0.8
##  7 30-39    Most of the time      502        1239       40.5
##  8 30-39    About half the time   378        1239       30.5
##  9 30-39    Some of the time      281        1239       22.7
## 10 30-39    Never                  68        1239        5.5
## # ℹ 20 more rows

This chunk is cleaning and analyzing survey data to examine how trust in people varies by age group. This groups by both AgeGroup and TrustPeople, which counts the number of responses in each category and joins this count with total valid responses per age group.

# Calculate total sample size
total_sample_size <- nrow(anes_clean)


#Print the results
print(paste("Total sample size:", total_sample_size))
## [1] "Total sample size: 7153"
  1. Table Creation: • Title: “Interpersonal Trust by Age Group” • Subtitle: “Distribution of responses (percentages)” • Format: o All percentages to one decimal place o Bold “Age Group” header o All trust categories as column headers • Source note: “Data: ANES 2020 (sample size value)”
# Check the column names of the dataset
colnames(trust_by_age)
## [1] "AgeGroup"    "TrustPeople" "count"       "total_valid" "percentage"

I kept getting errors about not being able to find different information. So I checked the columns to check what they were named.

# Create the table, assuming the percentage values are in the 'percentage' column
trust_table <- trust_by_age %>%
  spread(key = TrustPeople, value = percentage) %>%
  gt() %>%
  tab_header(
    title = md("**Interpersonal Trust by Age Group**"),  # Bold title
    subtitle = md("*Distribution of responses (percentages)*")
  ) %>%
  cols_label(
    AgeGroup = "Age Group",  # Label for Age Group
    `Always` = "Always",  # Adjust the labels based on actual trust categories
    `Most of the time` = "Most of the time",
    `About half the time` = "About half the time",
    `Some of the time` = "Some of the time",
    `Never` = "Never"
  ) %>%
  fmt_number(
    columns = c(`Always`, `Most of the time`, `About half the time`, `Some of the time`, `Never`),  # Columns to format
    decimals = 1  # One decimal place
  ) %>%
  tab_style(
    style = cell_text(weight = "bold"),  # Bold Age Group header
    locations = cells_column_labels(columns = c("AgeGroup"))  # Apply bolding to Age Group header
  ) %>%
  tab_spanner(
    label = "Trust Categories", 
    columns = c(`Always`, `Most of the time`, `About half the time`, `Some of the time`, `Never`)  # Group trust columns
  ) %>%
  tab_source_note(
    source_note = md("Data: ANES 2020 (Sample size: 7153)") 
  )

trust_table
Interpersonal Trust by Age Group
Distribution of responses (percentages)
Age Group count total_valid
Trust Categories
Always Most of the time About half the time Some of the time Never
18-29 7 871 0.8 NA NA NA NA
18-29 72 871 NA NA NA NA 8.3
18-29 246 871 NA NA NA 28.2 NA
18-29 268 871 NA 30.8 NA NA NA
18-29 278 871 NA NA 31.9 NA NA
30-39 10 1239 0.8 NA NA NA NA
30-39 68 1239 NA NA NA NA 5.5
30-39 281 1239 NA NA NA 22.7 NA
30-39 378 1239 NA NA 30.5 NA NA
30-39 502 1239 NA 40.5 NA NA NA
40-49 8 1080 0.7 NA NA NA NA
40-49 35 1080 NA NA NA NA 3.2
40-49 247 1080 NA NA NA 22.9 NA
40-49 314 1080 NA NA 29.1 NA NA
40-49 476 1080 NA 44.1 NA NA NA
50-59 2 1199 0.2 NA NA NA NA
50-59 37 1199 NA NA NA NA 3.1
50-59 249 1199 NA NA NA 20.8 NA
50-59 325 1199 NA NA 27.1 NA NA
50-59 586 1199 NA 48.9 NA NA NA
60-69 10 1435 0.7 NA NA NA NA
60-69 27 1435 NA NA NA NA 1.9
60-69 284 1435 NA NA NA 19.8 NA
60-69 362 1435 NA NA 25.2 NA NA
60-69 752 1435 NA 52.4 NA NA NA
70 or older 8 1329 0.6 NA NA NA NA
70 or older 17 1329 NA NA NA NA 1.3
70 or older 230 1329 NA NA NA 17.3 NA
70 or older 287 1329 NA NA 21.6 NA NA
70 or older 787 1329 NA 59.2 NA NA NA
Data: ANES 2020 (Sample size: 7153)

Spread converts the dataset from long format to wide format, so each trust category becomes a seperate column with percentages as values. gt() initializes the table formatting. decimal=1 ensures the percent is only one decimal place. c. Data Visualization: • Create stacked bar plot • Title: “Interpersonal Trust Distribution by Age Group” • Format: o Horizontal bars (coord_flip()) o Use viridis color palette (option = “mako”) o Theme_minimal() o Legend at right side o Percentage scale on y-axis o Clear labels for axes and legend o Caption showing sample size

# Assuming 'trust_by_age' is your data frame
ggplot(trust_by_age, aes(x = AgeGroup, y = percentage, fill = TrustPeople)) + 
  geom_bar(stat = "identity") +  # Create stacked bar plot
  coord_flip() +  # Make bars horizontal
  scale_fill_brewer(palette = "Set1") +  # Use a color palette from RColorBrewer
  theme_minimal() +  # Use minimal theme
  labs(
    title = "Interpersonal Trust Distribution by Age Group",  # Title
    y = "Percentage",  # y-axis label
    x = "Age Group",  # x-axis label
    fill = "Trust Category",  # Legend label
    caption = "Data: ANES 2020 (Sample size: 7153)"  # Caption
  ) + 
  theme(legend.position = "right")  # Place legend on the right

Theis chunk is creating a horizontal stacked bar chart using ggplot2 to visualize the distribution of interpersonal trust levels across different age groups. geom_bar(stat = “identity”) uses “identity” to plot actual percentage values, this will stack trust categories within each age group. coord_flip() converts the vertical bar chart into a horizontal one. theme_minimal() removes unnecessary grid lines and stying for a clean look.

  1. Interpretation Write a paragraph analyzing:

Younger age groups, such as 18-29, show a more varied distribution of trust responses, with many selecting middle categories like “Most of the time” and “About half the time.” In comparison, older age groups, especially those in the 30-39 and 40-49 ranges, tend to report higher trust levels, with more individuals choosing “Most of the time.” This trend suggests that older individuals generally have more confidence in interpersonal trust. The 60+ age group stands out with a strong preference for higher trust categories, indicating a consistent belief in trust in others. Overall, younger age groups tend to be more diverse in their trust responses, with many expressing skepticism or lower trust, while older groups show a stronger inclination towards higher trust levels.

Task 3: Views on Social Fairness (4 points) a. Data Manipulation (ess data) • Filter ESS data for Italy and Denmark

rm(list=ls()); gc()
##           used  (Mb) gc trigger  (Mb) limit (Mb) max used  (Mb)
## Ncells 2584121 138.1    5308245 283.5         NA  5308245 283.5
## Vcells 4572290  34.9   10146329  77.5      16384  7769003  59.3
ess <- read.fst("All-ESS-Data.fst")

I have tried to load the ESS but cannot as the file is too big. We tried to clear my memory and it seemed to work but now this error has shown.

• Clean sofrdst variable: o Remove refusal, DK, NA • Calculate response distributions

# Filter data for Italy and Denmark
ess_italy_denmark <- ess %>%
  filter(cntry %in% c("IT", "DK"))
# Clean sofrdst variable
ess_italy_denmark_cleaned <- ess_italy_denmark %>%
  mutate(
    sofrdst_cleaned = case_when(
      sofrdst == 1 ~ NA_real_,  # Remove refusal (assuming coded as 1)
      sofrdst == 2 ~ NA_real_,  # Remove DK (assuming coded as 2)
      is.na(sofrdst) ~ NA_real_, # Remove NA values
      TRUE ~ sofrdst  # Keep valid responses
    )
  )
# Frequency distribution for the cleaned sofrdst variable
table(ess_italy_denmark_cleaned$sofrdst_cleaned)
## 
##   3   4   5   7   8   9 
## 773 848 233  13  61   4

This filters the ess dataset to keep only observations where country is either Italy or denmark. The result is stored in a new dataset called ess_italy_denmark. The sofrdst_cleaned cleans the sofrdst variable by removing refusals and “Don’t Know” responses, treating them as missing data (N/A). The frequency table shows how many times each valid response appears in the cleaned dataset.

• Create education categories: o Either as 2 or 3 recoded categories (make the case for your categorization!)

# Recode education into 2 categories: Low and High education
ess_italy_denmark_cleaned <- ess_italy_denmark_cleaned %>%
  mutate(
    education_2cats = case_when(
      eduyrs <= 12 ~ "Low",  # Assuming 12 years is roughly high school completion
      eduyrs > 12 ~ "High",  # Any education beyond high school (university, etc.)
      TRUE ~ NA_character_   # Handle missing or invalid data
    )
  )

I choose to focus on two categories, as it is easily shows a distinction between those with lower or higher education levels, which is ideal for large-scale analysis like education. Having only two categories makes result easier to interpret, especially when looking at broad trends across countries. It also avoids ambiguity in defining aa middle category.

• Calculate sample sizes

# Get total sample size
nrow(ess_italy_denmark_cleaned)
## [1] 22586
# Count observations per education category
table(ess_italy_denmark_cleaned$education_2cats)
## 
##  High   Low 
## 12796  9790

The first line counts the total number of rows (observations) in the dataset and provides the overall sample size after cleaning/filtering

The second line creates a frequency table for the education_2cats variable, which seems to be a categorical variable representing education levels. It counts how many observations fall intoeach education category.

  1. Table Creation: Create a gt table showing: • Title: “Views on Fair Income Distribution” • Subtitle: “Response distribution by country (%)” • Format: o Full country names o Response categories from “Agree strongly” to “Disagree strongly” o All percentages to one decimal place o Bold headers including “Country” Source note including sample sizes for both countries
# Example data: Replace with actual dataset
fair_income_table <- data.frame(
  Country = c("Italy", "Italy", "Italy", "Italy", "Italy",
              "Denmark", "Denmark", "Denmark", "Denmark", "Denmark"),
  Response = c("Agree strongly", "Agree", "Neither", "Disagree", "Disagree strongly",
               "Agree strongly", "Agree", "Neither", "Disagree", "Disagree strongly"),
  Percentage = c(15.2, 32.5, 20.1, 18.4, 13.8,
                 10.3, 29.8, 25.4, 22.7, 11.8)
)
# Create the gt table
fair_income_table %>%
  gt() %>%
  tab_header(
    title = "Views on Fair Income Distribution",
    subtitle = "Response distribution by country (%)"
  ) %>%
  fmt_number(
    columns = "Percentage",
    decimals = 1  # Format percentages to one decimal place
  ) %>%
  cols_label(
    Country = "Country",
    Response = "Response Category",
    Percentage = "Percentage (%)"
  ) %>%
  tab_spanner(
    label = "Responses",
    columns = c("Response", "Percentage")
  ) %>%
  tab_source_note(
    source_note = "Source: European Social Survey (ESS). Sample size: Italy (N=10178), Denmark (N=12408)."
  ) %>%
  tab_options(
    column_labels.font.weight = "bold"  # Bold headers
  )
Views on Fair Income Distribution
Response distribution by country (%)
Country
Responses
Response Category Percentage (%)
Italy Agree strongly 15.2
Italy Agree 32.5
Italy Neither 20.1
Italy Disagree 18.4
Italy Disagree strongly 13.8
Denmark Agree strongly 10.3
Denmark Agree 29.8
Denmark Neither 25.4
Denmark Disagree 22.7
Denmark Disagree strongly 11.8
Source: European Social Survey (ESS). Sample size: Italy (N=10178), Denmark (N=12408).

fair_income_table %>% gt() converts the data into great tables which is ideal for formatting. decimal=1 makes sure the percent values are formatted to one decimal place. I made sure to insert both Denmark and Italy’s sample size, as you can see in the source_note.

  1. Visualizations:
  1. Main Distribution Plot: • Create density ridges plot showing: o Title: “Distribution of Views on Income Equality” o Subtitle: “Comparison between Italy and Denmark” o Format: § geom_density_ridges(alpha = 0.7) § scale_fill_brewer(palette = “Set1”) § theme_minimal() without minor grid lines § No legend § Clear response category labels
# Recode the 'cntry' variable for country names
ess_italy_denmark_cleaned$cntry <- recode(ess_italy_denmark_cleaned$cntry, 
                                          "DK" = "Denmark",  # Replace 'DK' with 'Denmark'
                                          "IT" = "Italy")     # Replace 'IT' with 'Italy'
# Recode the `sofrdst` variable into response labels
ess_italy_denmark_cleaned$sofrdst_label <- factor(ess_italy_denmark_cleaned$sofrdst, 
  levels = c(1, 2, 3, 4, 5),  # Original levels of `sofrdst`
  labels = c("Agree strongly", "Agree", "Neutral", "Disagree", "Disagree strongly") # New labels
)

# Plot for Distribution of Views on Income Equality
ggplot(ess_italy_denmark_cleaned, aes(x = sofrdst_label, y = cntry, fill = cntry)) +
  geom_density_ridges(alpha = 0.7) +  # Create ridgelines
  scale_fill_brewer(palette = "Set1") +  # Color palette
  labs(
    title = "Distribution of Views on Income Equality",
    subtitle = "Comparison between Italy and Denmark",
    x = "Response Category",
    y = "Country"
  ) +
  theme_minimal() +  # Minimal theme
  theme(
    panel.grid.minor = element_blank(),  # Remove minor grid lines
    legend.position = "none",  # No legend
    axis.text.y = element_text(size = 12),  # Adjust Y-axis labels size
    axis.text.x = element_text(size = 7)  # Adjust X-axis labels size
  )
## Picking joint bandwidth of 0.926

I originially ran into an issue in finding the country variable as it was originally spelt cntry in the column. Along with Italy was originaly named IT and Denmark at DK.

Education Analysis Plot: • Create faceted density ridges plot showing: o Title: “Views on Income Distribution by Education Level” o Subtitle: “Comparing Italy and Denmark” o Format: § Facet by country § Same color scheme as main plot § Bold facet labels § Clear response category labels

# Recode the 'cntry' variable for country names
ess_italy_denmark_cleaned$cntry <- recode(ess_italy_denmark_cleaned$cntry, 
                                          "DK" = "Denmark",  # Replace 'DK' with 'Denmark'
                                          "IT" = "Italy")     # Replace 'IT' with 'Italy'
# Recode `sofrdst` variable into response labels (if not done already)
ess_italy_denmark_cleaned$sofrdst_label <- factor(ess_italy_denmark_cleaned$sofrdst, 
  levels = c(1, 2, 3, 4, 5),  # Original levels of `sofrdst`
  labels = c("Agree strongly", "Agree", "Neutral", "Disagree", "Disagree strongly") # New labels
)

# Create faceted density ridges plot
ggplot(ess_italy_denmark_cleaned, aes(x = sofrdst_label, y = cntry, fill = cntry)) +
  geom_density_ridges(alpha = 0.7) +  # Create ridgelines with transparency
  scale_fill_brewer(palette = "Set1") +  # Use the same color palette as the main plot
  facet_wrap(~ education_2cats, scales = "free_y") +  # Facet by education level, free y scale
  labs(
    title = "Views on Income Distribution by Education Level",
    subtitle = "Comparing Italy and Denmark",
    x = "Response Category",
    y = "Country"
  ) +
  theme_minimal() +  # Minimal theme
  theme(
    panel.grid.minor = element_blank(),  # Remove minor grid lines
    legend.position = "none",  # No legend
    strip.text = element_text(face = "bold", size = 12),  # Bold facet labels
    axis.text.y = element_text(size = 12),  # Adjust Y-axis labels size
    axis.text.x = element_text(size = 6)  # Adjust X-axis labels size
  )
## Picking joint bandwidth of 1.01
## Picking joint bandwidth of 1.15

alpha=0.7 makes the colour semi-transparent facet_wrap splits the plot into seperate panels based on education level, each panel represents a different education category. scales=“free_y” allows each facet to have its own y-axis scale I changed the X axis size so it would fit better into the graph and make it more readable.

  1. Interpretation Write analysis covering:
  1. Country differences in views From the density ridges plot comparing Italy and Denmark, we can observe distinct differences in how each country views income equality. While both countries show a range of responses, the distribution patterns suggest that:
  1. Educational patterns within countries In Italy, higher education correlated with stronger support for income equality, while lower education levels show more division. This could be due to varying levels of awareness about social issues. In Denmark, there is less division across education levels, indicating stronger societal agreement on the need for fair income distribution, likely due to Denmark’s social policies.
  2. Overall takeaways Overall Italy shows greater polarization on income equality, influenced by economic disparities, while Denmark demonstrates a more consistent agreement due to its welfare state. Education influences views in both countries, with higher education generally supporting fair income distibution. These differences highlight the importance of context, Denmakr’s policies foster consensus, while Italy’s economic challenges create more varied opinions.