In Pecners/cityforwardcollective:

knitr::opts_chunk$set(echo = FALSE, message = FALSE, warning = FALSE)
library(tidyverse)
library(knitr)
library(kableExtra)
library(RColorBrewer)
library(scales)
library(wisconsink12)
library(cityforwardcollective)
library(googlesheets4)

# Avoid Google API Limits by saving local
# copy and refreshing only after 1 hour

if (!dir.exists("../data")) {

  dir.create("../data")

  gs4_auth(email = "stcmassessments@gmail.com")
  raw_results <- read_sheet(params$survey_url)

  last_data <- list("d" = raw_results, last_refresh = Sys.time())
  saveRDS(last_data, "../data/last_data.rds")
} else {

  last_data <- readRDS("../data/last_data.rds")

}

if (Sys.time() - last_data$last_refresh > 60) {

  gs4_auth(email = "stcmassessments@gmail.com")
  raw_results <- read_sheet(params$survey_url)

} else {

  raw_results <- last_data$d

}

raw_results <- raw_results %>%

  # Add params filters here

  filter()

Aggregate Responses (\hl{\textbf{n = `r nrow(raw_results)`}})

\newpage

Respondents

responses <- raw_results %>%
  select(16, 18)


names(responses) <- c("role", "sector")

responses <- responses %>%
           mutate(sector = str_replace_all(sector, " School", ""))


responses %>%
  group_by(role, sector) %>%
  summarise(value = n()) %>%
  modify_at("sector", factor, levels = c(map_chr(1:length(unique(responses$sector)), function(x) responses %>%
                                                                              group_by(sector) %>%
                                                                              summarise(tot = n()) %>%
                                                                              arrange(tot) %>%
                                                                              .$sector %>%
                                                                              .[x]))) %>%
  ggplot(aes(fill = sector, values = value)) +
  geom_waffle(color = "white", radius = unit(3, "pt"), size = 2, ) +
  theme(line = element_blank(),
        axis.text = element_blank(),
        legend.position = "bottom",
        legend.text = element_text(),
        panel.spacing.x = unit(1, "lines"),) +
  facet_wrap(~ role, nrow = 1, strip.position = "bottom", 
             labeller = label_wrap_gen(width = 15, multi_line = TRUE)) +
  coord_equal() +
  labs(fill = "", title = "Response Breakdown by Role and Sector",
       subtitle = "Each square is one respondent") +
  scale_fill_discrete(labels = function(x) str_wrap(x, 20)) +
  guides(fill = guide_legend(nrow = 2))

Logistics

responses <- raw_results %>%
  select(-c(1:2)) %>%
  modify_at(1:2, factor, levels = c("Very Poor",
                                    "Poor",
                                    "Fair",
                                    "Good",
                                    "Very Good"), ordered = TRUE) %>%
  modify_at(c(3:9), factor, levels = c("Strongly Disagree",
                                                   "Disagree",
                                                   "Unsure",
                                                   "Agree",
                                                   "Strongly Agree"), ordered = TRUE)


logistics <- responses %>%
  select(1:2) %>%
  gather(., key = "questions", value = "rating") %>%
  mutate(questions = str_replace_all(str_extract(questions, pattern = "\\[.*\\]$"), "\\[|\\]", "")) %>%
  modify_at(2, factor, levels = c("Very Poor",
                                  "Poor",
                                  "Fair",
                                  "Good",
                                  "Very Good"))

logistics %>%
  ggplot(aes(questions, fill = rating)) +
  geom_bar(position = "stack", width = .5) +
  theme_minimal() +
  scale_fill_brewer(palette = "RdYlGn", drop = FALSE) +
  labs(x = "", y = "Response Count", fill = "",
       title = "Logistics Feedback") +
  coord_flip()

General Session Feedback

gen_ses <- responses %>%
  select(3:7) %>%
  gather(., key = "questions", value = "rating") %>%
  mutate(questions = str_replace_all(str_extract(questions, pattern = "\\[.*\\]$"), "\\[|\\]", "")) %>%
  modify_at(2, factor, levels = c("Strongly Disagree",
                                  "Disagree",
                                  "Unsure",
                                  "Agree",
                                  "Strongly Agree"))

gen_ses %>%
  ggplot(aes(questions, fill = rating)) +
  geom_bar(position = "stack", width = .5) +
  theme_minimal() +
  scale_fill_brewer(palette = "RdYlGn", drop = FALSE) +
  labs(x = "", y = "Response Count", fill = "",
       title = "This session...") +
  coord_flip() +
  scale_x_discrete(labels = function(questions) str_wrap(questions, width = 20))

Facilitator Feedback

fac <- responses %>%
  select(8:9) %>%
  gather(., key = "questions", value = "rating") %>%
  mutate(questions = str_replace_all(str_extract(questions, pattern = "\\[.*\\]$"), "\\[|\\]", "")) %>%
  modify_at(2, factor, levels = c("Strongly Disagree",
                                  "Disagree",
                                  "Unsure",
                                  "Agree",
                                  "Strongly Agree"))

fac %>%
  ggplot(aes(questions, fill = rating)) +
  geom_bar(position = "stack", width = .5) +
  theme_minimal() +
  scale_fill_brewer(palette = "RdYlGn", drop = FALSE) +
  labs(x = "", y = "Response Count", fill = "",
       title = "The facilitator(s)...") +
  coord_flip() +
  scale_x_discrete(labels = function(questions) str_wrap(questions, width = 20)) 

comments <- responses %>%
  select(10) %>%
  filter(!is.na(.)) %>%
  kbl(booktabs = T, linesep = "") %>%
  kable_styling(latex_options = c("striped", "HOLD_position")) %>%
  column_spec(1, width = "40em")

comments

\newpage

Open-ended Feedback

\renewcommand{\arraystretch}{1.5}

for (i in c(11:13)) {
oe_responses <- responses %>%
  select(i) %>%
  filter(. != "NA" & . != "(left blank)") %>%
  kbl(booktabs = T, linesep = "") %>%
  kable_styling(latex_options = c("striped", "HOLD_position")) %>%
  column_spec(1, width = "40em")

print(oe_responses)
}