Top 20 Responses to Key Questions in the 2024 Stack Overflow Survey

Analysis of AI Tool Usage and Work Situation Across Countries

Figure 1: The Sankey diagram illustrates the top 20 responses to key questions in the 2024 Stack Overflow survey, tracking the flow of responses about AI tool usage and work situations in selected countries. The responses are divided into categories like “Yes,” “No Plan,” “Plan Soon,” “Hybrid,” “In-person,” and “Remote,” and the paths are color-coded to show different questions and responses from each country.

Steps to Create this Graphic

1. Load Packages & Setup

pacman::p_load(
  tidyverse,     # Easily Install and Load the 'Tidyverse'
  ggtext,        # Improved Text Rendering Support for 'ggplot2'
  showtext,      # Using Fonts More Easily in R Graphs
  janitor,       # Simple Tools for Examining and Cleaning Dirty Data
  skimr,         # Compact and Flexible Summaries of Data
  scales,        # Scale Functions for Visualization
  lubridate,     # Make Dealing with Dates a Little Easier
  MetBrewer,     # Color Palettes Inspired by Works at the Metropolitan Museum of Art
  MoMAColors,    # Color Palettes Inspired by Artwork at the Museum of Modern Art in New York City
  glue,          # Interpreted String Literals
  ggalluvial,    # Alluvial Plots in 'ggplot2' # Alluvial Plots in 'ggplot2' # Alluvial Plots in 'ggplot2'
  ggrepel,       # Automatically Position Non-Overlapping Text Labels with 'ggplot2'
  magick         # Advanced Graphics and Image-Processing in R 
 )  

### |- figure size ----
camcorder::gg_record(
  dir    = here::here("temp_plots"),
  device = "png",
  width  =  7.77,
  height =  8,
  units  = "in",
  dpi    = 320
)

### |- resolution ----
showtext_opts(dpi = 320, regular.wt = 300, bold.wt = 800)

2. Read in the Data

tt <- tidytuesdayR::tt_load(2024, week = 36) 

response_crosswalk <- tt$qname_levels_single_response_crosswalk |> clean_names() |> glimpse()
survey_questions   <- tt$stackoverflow_survey_questions |> clean_names() |> glimpse()
single_response    <- tt$stackoverflow_survey_single_response |> clean_names() |> glimpse()

tidytuesdayR::readme(tt)
rm(tt)

3. Examing the Data

glimpse(response_crosswalk)
glimpse(survey_questions)
glimpse(single_response)

4. Tidy Data

# Tidyy
tidy_data <- single_response |>
    pivot_longer(cols = -c(response_id, country, currency, comp_total, converted_comp_yearly), 
                 names_to = "qname", 
                 values_to = "response_code") |>
    left_join(response_crosswalk, by = c("qname", "response_code" = "level")) |>
    left_join(survey_questions, by = "qname") |> 
    mutate(
        country = case_when(
            country == "United Kingdom of Great Britain and Northern Ireland" ~ "UK",
            country == "United States of America" ~ "USA",
            TRUE ~ as.factor(country)
        )
    ) |> 
    drop_na(country) 

# Prepare data for Alluvial plot
alluvial_data <- tidy_data |>
    filter(qname %in% c("remote_work", "ai_select")) |> 
    count(question, country, label) |>
    arrange(desc(n)) |>
    slice_head(n = 20) |> 
    mutate(
        question = str_remove(question, "\\s*\\*"),                             # Removing the '*' from questions
        question = case_when(
            question == "Do you currently use AI tools in your development process?" ~ "AI Tool Usage",
            question == "Which best describes your current work situation?" ~ "Work Situation",
            TRUE ~ question
        ),
        label = case_when(
            label == "No, and I don't plan to" ~ "No Plan",
            label == "No, but I plan to soon" ~ "Plan Soon",
            label == "Hybrid (some remote, some in-person)" ~ "Hybrid",
            TRUE ~ label
        ),
        ) |> 
    filter(!is.na(label))

# Plot data
plot_data <- alluvial_data |> 
    filter(!is.na(question))

5. Visualization Parameters

### |- plot aesthetics ----
bkg_col      <- colorspace::lighten('#f7f5e9', 0.05)    
title_col    <- "gray20"           
subtitle_col <- "gray20"     
caption_col  <- "gray30"   
text_col     <- "gray20"    
col_palette  <- MoMAColors::moma.colors(palette_name = "Koons", n = 2, type = 'discrete')

### |-  titles and caption ----
# icons
tt <- str_glue("#TidyTuesday: { 2024 } Week { 36 } &bull; Source: Stack Overflow Annual Developer Survey 2024<br>")
li <- str_glue("<span style='font-family:fa6-brands'>&#xf08c;</span>")
gh <- str_glue("<span style='font-family:fa6-brands'>&#xf09b;</span>")
mn <- str_glue("<span style='font-family:fa6-brands'>&#xf4f6;</span>")

# text
title_text    <- str_glue("Top 20 Responses to Key Questions in the 2024 Stack Overflow Survey")
subtitle_text <- str_glue("Analysis of AI Tool Usage and Work Situation Across Countries")
caption_text  <- str_glue("{tt} {li} stevenponce &bull; {mn} @sponce1(graphic.social) {gh} poncest &bull; #rstats #ggplot2")

### |-  fonts ----
font_add("fa6-brands", "fonts/6.4.2/Font Awesome 6 Brands-Regular-400.otf")
font_add_google("Oswald", regular.wt = 400, family = "title")
font_add_google("Merriweather Sans", regular.wt = 400, family = "subtitle")
font_add_google("Merriweather Sans", regular.wt = 400, family = "text")
font_add_google("Noto Sans", regular.wt = 400, family = "caption")
showtext_auto(enable = TRUE)

### |-  plot theme ----
theme_set(theme_minimal(base_size = 14, base_family = "text"))                

theme_update(
    plot.title.position   = "plot",
    plot.caption.position = "plot",
    legend.position       = 'top',
    plot.background       = element_rect(fill = bkg_col, color = bkg_col),
    panel.background      = element_rect(fill = bkg_col, color = bkg_col),
    plot.margin           = margin(t = 20, r = 20, b = 20, l = 20),
    axis.title.x          = element_text(margin = margin(10, 0, 0, 0), size = rel(1.1), color = text_col, family = "text", face = "bold", hjust = 0.5),
    axis.title.y          = element_text(margin = margin(0, 10, 0, 0), size = rel(1.1), color = text_col, family = "text", face = "bold", hjust = 0.5),
    axis.text             = element_text(size = rel(0.8), color = text_col, family = "text"),
    axis.line.x           = element_line(color = "gray40", linewidth = .15),
    panel.grid.minor.y    = element_blank(),
    panel.grid.major.y    = element_line(linetype = "dotted", linewidth = 0.1, color = 'gray'),
    panel.grid.minor.x    = element_blank(),
    panel.grid.major.x    = element_blank(),
)  

6. Plot

### |-  final plot ----  
p <- ggplot(plot_data, aes(axis1 = question, axis2 = country, axis3 = label, y = n)) +
    
    # Geoms
    geom_alluvium(aes(fill = question), width = 1/12) +
    geom_stratum(aes(fill = question), width = 1/8, alpha = 0.4) +
    geom_text(stat = "stratum", aes(label = after_stat(stratum)), size = 3.5, hjust = 0, nudge_x = 0.08) +
    
    # Labs
    labs(
        y = "Number of Responses",
        x = "",
        fill = "Question: ",
        title    = title_text,
        subtitle = subtitle_text,
        caption  = caption_text
    ) +
    
    # Scales
    scale_x_discrete(limits = c("Question", "Country", "Response"), expand = c(0.15, 0.05)) +
    scale_y_continuous(labels = scales::number_format(scale = 1/1000, suffix = " K")) +
    scale_fill_manual(values = col_palette, na.translate = FALSE) +    
    coord_cartesian(clip = 'off') +
    
    # Theme
    theme(
        plot.title = element_text(
            size = rel(1.3),
            family = "title",
            color = title_col,
            face = "bold",
            lineheight = 0.85,
            margin = margin(t = 5, b = 5)
        ),
        plot.subtitle = element_text(
            size = rel(1),
            family = "subtitle",
            color = title_col,
            lineheight = 1,
            margin = margin(t = 5, b = 15)
        ),
        plot.caption = element_markdown(
            size = rel(.5),
            family = "caption",
            color = caption_col,
            lineheight = 0.6,
            hjust = 0,
            halign = 0,
            margin = margin(t = 0, b = 0)
        )
    )

7. Save

### |-  plot image ----  
ggsave(
  filename = here::here("data_visualizations/TidyTuesday/2024/tt_2024_36.png"),
  plot = p,
  width  =  7.77,
  height =  8,
  units  = "in",
  dpi    = 320
)

### |-  plot thumbnail----  
magick::image_read(here::here("data_visualizations/TidyTuesday/2024/tt_2024_36.png")) |> 
  magick::image_resize(geometry = "400") |> 
  magick::image_write(here::here("data_visualizations/TidyTuesday/2024/thumbnails/tt_2024_36.png"))

8. Session Info

sessionInfo()

R version 4.4.1 (2024-06-14 ucrt)
Platform: x86_64-w64-mingw32/x64
Running under: Windows 10 x64 (build 19045)

Matrix products: default


locale:
[1] LC_COLLATE=English_United States.utf8 
[2] LC_CTYPE=English_United States.utf8   
[3] LC_MONETARY=English_United States.utf8
[4] LC_NUMERIC=C                          
[5] LC_TIME=English_United States.utf8    

time zone: America/New_York
tzcode source: internal

attached base packages:
[1] stats     graphics  grDevices datasets  utils     methods   base     

loaded via a namespace (and not attached):
 [1] htmlwidgets_1.6.4 compiler_4.4.0    fastmap_1.2.0     cli_3.6.3        
 [5] htmltools_0.5.8.1 tools_4.4.0       rstudioapi_0.16.0 yaml_2.3.10      
 [9] rmarkdown_2.28    knitr_1.48        jsonlite_1.8.8    xfun_0.47        
[13] digest_0.6.37     rlang_1.1.4       renv_1.0.3        evaluate_0.24.0