sample.qmd

---
title: "Demographics"
---

<!-- Setup -->

```{r}
source('setup.R')
```

`r nrow(survey_results)` respondents finished the survey, broken out by role and affiliation below:

::: {.panel-tabset}

### Role

```{r}
plotly_treemap_role_df <- survey_results |>  
  group_by(type = QID4, division = QID8) |> 
  summarise(n = n()) |> 
  mutate(
    type = ifelse(is.na(type), 'Unaffiliated', type),
    division = ifelse(is.na(division), 'Unaffiliated', division)
  )

plotly_treemap_role_df <- plotly_treemap_role_df |> 
  bind_rows(
    survey_results |> 
      group_by(division = QID4) |> 
      summarise(n = n()) |> 
      mutate(
        division = ifelse(is.na(division), 'Unaffiliated', division)
      )
  ) |> 
  mutate(ids = ifelse(is.na(type), division, paste0(type, "_", division))) |>  
  select(ids, everything()) |> 
  mutate(
    ids = ifelse(is.na(ids), "", ids),
    division = ifelse(is.na(division), type, division),
    type = ifelse(is.na(type), "", type),
    n = as.double(n)
  )

plot_ly(type = "treemap", 
        data = plotly_treemap_role_df,
        labels = ~division,
        parents = ~type,
        ids = ~ids,
        values = ~n,
        hoverinfo = "text",
        hovertemplate = "<b>%{parent}</b><br>Type: %{label}<br>Responses: %{value}<extra></extra>",
        textposition = "inside", # insidetextanchor = "middle",
        textfont = list(size = 14),
        branchvalues = "total"
        ) |> 
  layout(title = "Survey Respondents",
         plot_bgcolor  = background_color,
         paper_bgcolor = background_color)
```

### Affiliation

```{r}
plotly_treemap_affiliation_df <- survey_results |>  
  group_by(type = QID4, division = QID8) |> 
  summarise(n = n()) |> 
  mutate(
    type = ifelse(is.na(type), 'Unaffiliated', type),
    division = ifelse(is.na(division), 'Unaffiliated', division)
  )

plotly_treemap_affiliation_df <- plotly_treemap_affiliation_df |> 
  bind_rows(
    survey_results |> 
      group_by(type = QID8) |> 
      summarise(n = n()) |> 
      mutate(
        type = ifelse(is.na(type), 'Unaffiliated', type)
      )
  ) |> 
  mutate(ids = ifelse(is.na(division), type, paste0(division, "_", type))) |>  
  select(ids, everything()) |> 
  mutate(
    ids = ifelse(is.na(ids), "", ids),
    division = ifelse(is.na(division), "", division),
    type = ifelse(is.na(type), "", type),
    n = as.double(n)
  )

plot_ly(type = "treemap", 
        data = plotly_treemap_affiliation_df,
        labels = ~type,
        parents = ~division,
        ids = ~ids,
        values = ~n,
        hoverinfo = "text",
        hovertemplate = "<b>%{parent}</b><br>Type: %{label}<br>Responses: %{value}<extra></extra>",
        textposition = "inside", insidetextanchor = "middle",
        textfont = list(size = 14),
        branchvalues = "total"
        ) |> 
  layout(title = "Survey Respondents",
         plot_bgcolor  = background_color,
         paper_bgcolor = background_color)
```

:::

<b>`r round((sum(survey_results$QID22 == 'Yes', na.rm = TRUE) / nrow(survey_results) * 100))`%</b> of respondents said they have contributed to open source projects, either academically or personally.

```{r}
c1_df <- survey_results |> 
  mutate(QID4 = ifelse(is.na(QID4), 'Unafilliated', QID4)) |> 
  rename(`Respondent Type` = QID4) |> 
  group_by(QID22, `Respondent Type`) |> 
  summarise(Count = n(),
            Percent = n() / nrow(survey_results))

c1_df |>
  plot_ly(
    x = ~ QID22,
    y = ~ Percent,
    color = ~ `Respondent Type`,
    colors = viridis_pal(option = "D")(length(c1_df$`Respondent Type`))
  ) |>
  add_bars() |>
  layout(
    barmode = 'stack',
    plot_bgcolor = background_color,
    paper_bgcolor = background_color,
    xaxis = list(title = 'Have contributed to open source projects?'),
    yaxis = list(
      zerolinecolor = '#ffff',
      zerolinewidth = 2,
      gridcolor = 'ffff',
      tickformat = ".1%"
    )
  )
```

### Faculty and Staff

Of these respondents, `r nrow(survey_results[survey_results$QID4 == 'Faculty'|survey_results$QID4 == 'Staff',])` identified as faculty or staff (`r round(nrow(survey_results[survey_results$QID4 == 'Faculty'|survey_results$QID4 == 'Staff',]) / nrow(survey_results), 2) * 100`% of respondents). 

Faculty and staff respondents were distributed in tenure (years served) at the university as below:

```{r}
tenure_gg <- survey_results |> 
  mutate(QID6 = factor(QID6, 
                       levels = c('0-1 years', '1-3 years', '4-7 years', '7-10 years', '10+ years'))) |> 
  filter(QID4 %in% c('Faculty', 'Staff')) |> 
  ggplot(aes(y = QID6)) +
  geom_bar(fill = primary_color
           ) +
  scale_y_discrete(limits = rev) +
  labs(
    y = 'Years Served',
    x = 'Respondents'
  ) +
  facet_wrap(vars(QID4)) +
  bar_coord_flip_theme

ggplotly(tenure_gg, tooltip = 'count')
```

### Students

`r nrow(survey_results[survey_results$QID4 == 'Graduate Student'|survey_results$QID4 == 'Undergraduate Student',])` respondents identified at students (`r round(nrow(survey_results[survey_results$QID4 == 'Graduate Student'|survey_results$QID4 == 'Undergraduate Student',]) / nrow(survey_results), 2) * 100`% of respondents). Students came from degree programs in the following subjects:

```{r}
major_gg <- survey_results |> 
  filter(QID4 %in% c('Graduate Student', 'Undergraduate Student')) |> 
  mutate(
    QID4 = ifelse(QID4 == 'Graduate Student', 'Graduate', 'Undergraduate'),
    QID7 = str_to_title(QID7),
    QID7 = str_replace_all(QID7, '&', 'And'),
    QID7 = str_remove_all(QID7, ' Phd')) |>
  # Manually clean up some majors
  mutate(
    QID7 = case_when(
      QID7 == 'Ag And Applied Economics' ~ 'Applied And Agricultural Economics',
      QID7 == 'Communication Sciences And Disorders (Csd)' ~ 'Communication Sciences And Disorders',
      QID7 == 'Computer Sciences' ~ 'Computer Science',
      QID7 == 'Industrial And Systems Engineering Phd' ~ 'Industrial & Systems Engineering',
      QID7 == 'Math' ~ 'Mathematics',
      QID7 == 'Nurtritional Sciences' ~ 'Nutrition Science',
      is.na(QID7) ~ 'Not Provided',
      TRUE ~ QID7
    )
  ) |> 
  ggplot(aes(y = reorder(QID7, QID7,
                     function(x) length(x)))) +
  geom_bar(fill = primary_color) +
  scale_x_continuous(breaks = pretty_breaks()) +
  labs(
    y = 'Major',
    x = 'Respondents'
  ) +
  facet_wrap(vars(QID4)) +
  bar_coord_flip_theme

ggplotly(major_gg, tooltip = 'count')
```