index.Rmd

---
title: ""
author: ""
date: "`r format(Sys.time(), '%B %d, %Y %H:%M')`"
output:
  html_document:
    number_sections: false
    self_contained: TRUE
    code_folding: hide
    toc: TRUE
    toc_float: TRUE
    css: www/web_report.css
    editor_options:
      chunk_output_type: console
---

<link rel="stylesheet" href="//fonts.googleapis.com/css?family=Lato" />

![](www/images/urban-institute-logo.png)


```{r rmarkdown-setup, echo = FALSE}
knitr::opts_chunk$set(warning = FALSE)
knitr::opts_chunk$set(message = FALSE)
```



# RAADS Grad School Survey

This year, we sent around a brief survey to better inform the annual RAADS grad school panel. The reults of the survey are presented below.

```{r setup}
options(scipen = 999)

library(tidyverse)
library(urbnthemes)

set_urbn_defaults("print")

# load data
source("code/analyze-survey-data.R")

```


### Degrees
 
Of the four degrees listed on the survey, `r tolower((filter(response_data, question == "degree") %>% arrange(desc(n)) %>% pull(response))[1])` was the most common degree selected. There was one response outside the set answers, expresseing interest in learning about certificates that could be earned while working at Urban.

```{r degree-plot}

response_data %>% 
  filter(question == "degree") %>% 
  ggplot(mapping = aes(x = reorder(response, desc(n)), y = n)) +
  geom_col() +
  geom_text(mapping = aes(label = n),
            vjust = -1, size = 4) +
  scale_y_continuous(expand = expand_scale(mult = c(0, .1))) +
  urbnthemes::remove_axis() +
  labs(x = NULL, y = NULL,
       title = "What kind of degree(s) are you interested in?")


```

### Fields

`r (filter(response_data, question == "field") %>% arrange(desc(n)) %>%  pull(response))[1]` was the most common field RAADS were interested in. `r filter(response_data, response == (filter(response_data, question == "field") %>% arrange(desc(n)) %>%  pull(response))[1]) %>% pull(n)` of `r mean(response_data$total)` total respondents selected it. Second and third place went to `r (filter(response_data, question == "field") %>% arrange(desc(n)) %>%  pull(response))[2]` and `r (filter(response_data, question == "field") %>% arrange(desc(n)) %>%  pull(response))[3]`.


```{r field-plot}

annotation_text <- paste0(mean(response_data$total),
                          " total \n respondents")

response_data %>% 
  filter(question == "field") %>% 
  ggplot(mapping = aes(x = reorder(response, desc(n)), y = n)) +
  geom_col() +
  geom_text(mapping = aes(label = n),
            hjust = -.5, size = 3) +
  scale_y_continuous(expand = expand_scale(mult = c(0, .1)),
                     limits = c(0, 42)) +
  geom_hline(aes(yintercept = total),
             linetype = "dotted") +
  annotate(geom = "text", x = "Math or statistics", 
           y = mean(response_data$total) - .5,
           label = annotation_text,
           hjust = 1) +
  coord_flip() +
  remove_axis(axis = "x", flip = TRUE) +
  labs(x = NULL, y = NULL,
       title = "What field(s) are you interested in?") +
  theme(plot.title = element_text(hjust = -.75))

```

### Process

The survey asked which parts of the process respondents would like advice on. Most RAADS indicated that they were looking for advice on `r tolower((filter(response_data, question == "process") %>% arrange(desc(n)) %>% pull(response))[1])`.

```{r process-plot}

response_data %>% 
  filter(question == "process") %>% 
  ggplot(mapping = aes(x = reorder(response, desc(n)), y = n)) +
  geom_col() +
  geom_text(mapping = aes(label = n),
            vjust = -1, size = 4) +
  scale_y_continuous(expand = expand_scale(mult = c(0, .1))) +
  urbnthemes::remove_axis() +
  labs(x = NULL, y = NULL,
       title = "What part of the process are you looking for advice on?")


```

### Other

The following comments were added in addition to the questions posed:

```{r other, results=FALSE}

# make vector of other responses
other <- c(
  data %>% 
  filter(!is.na(other),
         other != "nope") %>% 
  pull(other),
  "Financial trade-offs between stopping work and the value of the degree")

# this is to make sure we use the right number of bullets
length(other)

```
* `r other[1]`

* `r other[2]`

* `r other[3]`

* `r other[4]`

* `r other[5]`

* `r other[6]`

* `r other[7]`

* `r other[8]`