five_question_summary.R

# Libraries

library(dplyr)
library(tidyverse)
library(ggplot2)

# Load data
colnames <- c("time", "email", "name", "section", 
              "challenge", "support", 
              "competence", "autonomy", "relatedness",
              "elaborate", "needs")


## Change name of input file as needed
responses <- read_csv("put the name of your CSV here", 
                      skip = 1,
                      col_names = colnames)

# Summary stats 

summary(responses)


# Combined boxplots 

## Challenge and support variables
y_range <- range(c(responses$challenge, responses$support))
par(mfrow = c(1,2))
boxplot(responses$challenge, main = "Challenge", 
        col = "lightblue", border = "black", ylim = y_range)
boxplot(responses$support, main = "Support", 
        col = "lightgreen", border = "black", ylim = y_range)
par(mfrow = c(1,1))

## Self Determination variables 
y_range <- range(c(responses$competence, responses$autonomy, responses$relatedness))
par(mfrow = c(1,3))
boxplot(responses$competence, main = "Competence", 
        col = "lightblue", border = "black", ylim = y_range)
boxplot(responses$autonomy, main = "Autonomy", 
        col = "lightgreen", border = "black", ylim = y_range)
boxplot(responses$relatedness, main = "Relatedness", 
        col = "lightcoral", border = "black", ylim = y_range)
par(mfrow = c(1,1))


# The "bubble plot"

## Start with a basic scatterplot of support vs challenge
## Condense responses to a single data frame with unique rows plus frequencies
chall_support <- responses[, c(5,6)] |>
  group_by_all() |>
  summarise(freq = n())

## Scatterplot of responses with frequencies labelled 
chall_support |> ggplot(aes(challenge, support)) +
  geom_hline(yintercept = 3, linetype = "dashed", color="black", size=0.2) + 
  geom_vline(xintercept = 3, linetype = "dashed", color="black", size=0.2) +
  geom_point(color = "lightblue", size = chall_support$freq, alpha = 1) + 
  geom_text(aes(label = sprintf("%d", freq)), size = 5) + 
  labs(title = "Support vs Challenge", x = "Challenge", y = "Support") + 
  scale_x_continuous(limits = c(1,5)) + 
  scale_y_continuous(limits = c(1,5)) +
  coord_cartesian(clip = "off") + 
  theme(plot.margin = margin(1,1,1,1,"cm"))
  

## Trying it with SDT variables instead 

## Change the columns as needed
## 7 = competence
## 8 = autonomy
## 9 = relatedness

sdt_paired <- responses[, c(8,9)] |>
  group_by_all() |> 
  summarise(freq = n())

## Change variable names as needed
sdt_paired |> ggplot(aes(autonomy, relatedness)) +
  geom_hline(yintercept = 3, linetype = "dashed", color="black", size=0.2) + 
  geom_vline(xintercept = 3, linetype = "dashed", color="black", size=0.2) +
  geom_point(color = "lightblue", size = auto_related$freq, alpha = 1) + 
  geom_text(aes(label = sprintf("%d", freq)), size = 5) + 
  labs(title = "Autonomy vs Relatedness", x = "Autonomy", y = "Relatedness") + 
  scale_x_continuous(limits = c(1,5)) + 
  scale_y_continuous(limits = c(1,5)) +
  coord_cartesian(clip = "off") + 
  theme(plot.margin = margin(1,1,1,1,"cm"))