TRM Practical Data Analysis - Basic level
Thach Tran
2024-02-25
TRM Practical Data Analysis - Basic level
Lecture 3. Data visualisation
Task 1. Read the “Professorial Salaries” and name this dataset “salary”
salary = read.csv("C:\\Thach\\UTS\\Teaching\\TRM\\Practical Data Analysis\\2024_Autumn semester\\Data\\Professorial Salaries.csv")Libraries needed
library(ggplot2)
library(grid)
library(gridExtra)
library(dplyr)##
## Attaching package: 'dplyr'## The following object is masked from 'package:gridExtra':
##
## combine## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, unionlibrary(GGally)## Registered S3 method overwritten by 'GGally':
## method from
## +.gg ggplot2library(ggthemes)## Warning: package 'ggthemes' was built under R version 4.3.2Task 2. Distribution of numeric data
2a. Histogram to describe the distribution of professors’ salaries
p = ggplot(data = salary, aes(x = Salary))
p1 = p + geom_histogram(color = "white", fill = "blue")
p1## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
Change number of cases to density (y-axis) and add a normal density curve
p = ggplot(data = salary, aes(x = Salary))
p1 = p + geom_histogram(color = "white", fill = "blue")
p2 = p + geom_histogram(aes(y = ..density..), color = "white", fill = "blue")
p2 = p2 + geom_density(col="red")
grid.arrange(p1, p2, nrow = 2, top = textGrob("Distribution of professors' salaries by sex", gp = gpar(fontsize = 20, font = 1)))## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.## Warning: The dot-dot notation (`..density..`) was deprecated in ggplot2 3.4.0.
## ℹ Please use `after_stat(density)` instead.
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_lifecycle_warnings()` to see where this warning was
## generated.## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
2b. Histogram to describe the distribution of professors’ salaries by sex
p = ggplot(data = salary, aes(x = Salary, fill = Sex))
p1 = p + geom_histogram(position = "dodge")
p2 = ggplot(data = salary, aes(x = Salary, fill = Sex, color = Sex)) + geom_density(alpha = 0.1)
grid.arrange(p1, p2, nrow = 2, top = textGrob("Distribution of professors' salaries by sex", gp = gpar(fontsize = 20, font = 1)))## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
Task 3. Distribution of categorical data
3a. Barplot to describe the distribution of professors’ ranks
p = ggplot(data = salary, aes(x = Rank, fill = Rank, col = Rank))
p1 = p + geom_bar(position = "dodge")
salary$Prof.Rank = factor(salary$Rank, levels = c("AsstProf", "AssocProf", "Prof"))
p = ggplot(data = salary, aes(x = Prof.Rank, fill = Prof.Rank, col = Prof.Rank))
p2 = p + geom_bar(position = "dodge")
grid.arrange(p1, p2, nrow = 2, top = textGrob("Distribution of professors' rank", gp = gpar(fontsize = 20, font = 1)))
3b. Barplot to describe the distribution of professors’ ranks by sex
p = ggplot(data = salary, aes(x = Prof.Rank, fill = Sex, col = Sex))
p1 = p + geom_bar(position = "dodge")
p1 + ggtitle("Distribution of professors' rank by sex")
Add number (%) to the graphs
p = ggplot(salary %>% count(Prof.Rank, Sex) %>% mutate(pct = n/sum(n)), aes(factor(Prof.Rank), n, fill = Sex))
p = p + geom_bar(stat ="identity")
p = p + geom_text(aes(label = paste0(sprintf("%1.1f", pct*100),"%")), position = position_stack(vjust=0.5))
p + labs(x = "Professors' ranks", y = "Number of cases") + ggtitle("Professors' rank by sex")
Task 4. Comparison of numeric data
4a. Boxplot to describe the differences in professors’ salaries between males and females
p = ggplot(data = salary, aes(x = Sex, y = Salary, fill = Sex, col = Sex))
p1 = p + geom_boxplot(col = "black") + geom_jitter(alpha = 0.05)
p1 + labs(x = "Sex", y = "Salaries (USD)") + ggtitle("Professors' salaries by sex") + theme_bw()
4b. Boxplot to describe the differences in professors’ salaries among professors’ rank
p = ggplot(data = salary, aes(x = Prof.Rank, y = Salary, fill = Prof.Rank, col = Prof.Rank))
p1 = p + geom_boxplot(col = "black") + geom_jitter(alpha = 0.05)
p1 + labs(x = "Rank", y = "Salaries (USD)") + ggtitle("Professors' salaries by rank") + theme_bw()
Boxplot to describe the differences in professors’ salaries by rank and sex
p = ggplot(data = salary, aes(x = Prof.Rank, y = Salary, fill = Sex, col = Sex))
p1 = p + geom_boxplot(col = "black") + geom_jitter(alpha = 0.05)
p1 + labs(x = "Professors' Ranks", y = "Salaries (USD)") + ggtitle("Professors' salaries by rank and sex") + theme_bw()
Task 5. Correlation between numeric data
5a. Scatterplot to describe the correlation between professors’ salaries and their time in service
p = ggplot(data = salary, aes(x = Yrs.service, y = Salary))
p1 = p + geom_point() + geom_smooth() + labs(x = "Time in service (years)", y = "Professors' salaries (USD)") + ggtitle("Correlation between professors' salaries and time in service") + theme_bw()
p1## `geom_smooth()` using method = 'loess' and formula = 'y ~ x'
5b. Scatterplot to describe the correlation between professors’ salaries and their time in service by sex
p = ggplot(data = salary, aes(x = Yrs.service, y = Salary, fill = Sex, col = Sex))
p2 = p + geom_point() + geom_smooth(method = "lm", formula = y ~ x + I(x^2) + I(x^3)) + labs(x = "Time in service (years)", y = "Professors' salaries (USD)") + ggtitle("Correlation between professors' salaries and time in service by sex") + theme_bw()
p2