---
title: "Wellbeing and income re-analysis"
author: "John Wilcox"
date: "2024-12-25"
output: html_document
---
```{r setup, include=FALSE}

install.packages("rmarkdown")
install.packages('readr')
install.packages('tidyverse')
install.packages('writexl')
install.packages('readxl')
install.packages('ggplot2')
install.packages('Hmisc')
install.packages('MKmisc')
install.packages("corrplot")
install.packages("ggcorrplot")
install.packages("MKinfer")

# knitr::opts_chunk$set(echo = TRUE)
```


```{r setup 2, include=TRUE}
# knitr::opts_chunk$set(echo = TRUE)

library(rmarkdown)
library(readr)
library(tidyverse)
library(writexl)
library(readxl)
library(ggplot2)
library(Hmisc)
library(corrplot)
library(ggcorrplot)
library(MKinfer)

```

```{r}

#Setting seed for reproducibility

set.seed(123)

# Loading data from OSF repository here: https://osf.io/qye4a/

df1 = read_csv("C:/Users/john-/Downloads/Income_and_emotional_wellbeing_a_conflict_resolved.csv")


# Note: base of the log for the transformed data is approximately 2.718854203674
# Wellbeing = Mean response to “How do you feel right now?” on a continuous response scale with end points labeled “Very bad” and “Very good.”

# Returning range of incomes: range is $15k to $625k

range(df1$income)


# Produce wellbeing quantiles from data

quantile(df1$wellbeing, probs = c(.25,.5,.75,1))

# The quantiles are
#      25%      50%      75%     100% 
#  55.8050  63.1910  71.2945 100.0000 


#Make new wellbeing variables allocating individuals to four categories

df1 = df1%>%mutate(wellbeing_quart = case_when(wellbeing < 55.8050 ~ "Bottom",
                           55.8050 <= wellbeing & wellbeing < 63.1910 ~ "2",
                           63.1910 <= wellbeing & wellbeing < 71.2945 ~ "3",
                           71.2945 <= wellbeing ~ "Top"))


# Scatterplot of full data

ggplot(data = df1, aes(x = log_income, y = wellbeing)) +
        geom_jitter(width = 0.25, size = 2, alpha = 0.2)+
        geom_smooth(method = "lm")+
        scale_x_continuous(name = "Income", breaks=c(9.5,10, 10.5,11,11.5,12, 12.5,13, 13.5), labels = c("$13k","$22k","$36k","$60k","$99k","$163k","$269k","$443k", "$731k")) +
        labs(title = "The relationship between Wellbeing and Income",
             caption = "Wellbeing was measured as the mean response to 'How do you feel right now?' on a continuous response scale with end points labeled 'Very bad' and 'Very good'",
             y = "Wellbeing")+
        theme_classic()


# Scatterplot of random sample

ggplot(data = df1[sample(nrow(df1), 4000), ], aes(x = log_income, y = wellbeing)) +
        geom_jitter(width = 0.25, size = 2, alpha = 0.2)+
        geom_smooth(method = "lm") +
        scale_x_continuous(name = "Income", breaks=c(9.5,10, 10.5,11,11.5,12, 12.5,13, 13.5), labels = c("$13k","$22k","$36k","$60k","$99k","$163k","$269k","$443k", "$731k")) +
        scale_y_continuous(name = "Average response: 'How do you feel right now?'", breaks=c(0,25,50,75,100), labels = c("0 ('Very bad')","25","50","75","100 ('Very good')")) +  
        labs(title = "The relationship between Wellbeing and Income")+
        theme_minimal()



# Scatterplot for individuals past the $75k mark 

ggplot(data = df1%>%filter(log_income >11.5), aes(x = log_income, y = wellbeing)) +
        geom_jitter(width = 0.25, size = 2, alpha = 0.2)+
        geom_smooth(method = "lm") +
        scale_x_continuous(name = "Income", breaks=c(9.5,10, 10.5,11,11.5,12, 12.5,13, 13.5), labels = c("$13k","$22k","$36k","$60k","$99k","$163k","$269k","$443k", "$731k")) +
        scale_y_continuous(name = "Average response: 'How do you feel right now?'", breaks=c(0,25,50,75,100), labels = c("0 ('Very bad')","25","50","75","100 ('Very good')")) +  
        labs(title = "The relationship between Wellbeing and Income")+
        theme_minimal()


# Scatterplot for individuals reporting highest and lowest wellbeing past the $75k mark

ggplot(data = df1%>%filter(log_income >11.5,
                           wellbeing_quart != "2",
                           wellbeing_quart != "3"), aes(x = log_income, y = wellbeing, color = wellbeing_quart)) +
        geom_jitter(width = 0.25, size = 2, alpha = 0.2)+
        geom_smooth(method = "lm", fill = NA) +
        scale_x_continuous(name = "Income", breaks=c(9.5,10, 10.5,11,11.5,12, 12.5,13, 13.5), labels = c("$13k","$22k","$36k","$60k","$99k","$163k","$269k","$443k", "$731k")) +
        scale_y_continuous(name = "Average response: 'How do you feel right now?'", breaks=c(0,25,50,75,100), labels = c("0 ('Very bad')","25","50","75","100 ('Very good')")) +  
        labs(title = "The relationship between Wellbeing and Income")+
        theme_minimal()

```