Data Analysis YR.4 SEM1

BioMarin Pharmaceuticals Data Report 2024

Author

Joanna Zaremba, C00258740

BioMarin Website

Setting Up

library(tidyverse)
library(gghighlight)
library(knitr)
Warning: package 'knitr' was built under R version 4.4.2
employee <- read_csv("employee.csv")

The chunk above sets up the Quarto file by;

  1. Loading the tidyverse package.
  2. Rendering the “employee.csv” data base into RStudio and Quarto.

I am using the “employee.csv” file for this assignment.

Demographics of Age by Gender working at BioMarin Pharmaceuticals

summarize(employee, min(Age)) #minimum age#
# A tibble: 1 × 1
  `min(Age)`
       <dbl>
1         18
summarize(employee, max(Age)) #maximum age#
# A tibble: 1 × 1
  `max(Age)`
       <dbl>
1         60
employee <- mutate(employee, Age_Bracket = case_when(
  Age =18 & Age <21 ~ "18-20",
  Age =21 & Age <31 ~ "20-30",
  Age =31 & Age <41 ~ "30-40",
  Age =41 & Age <51 ~ "40-50",
  Age =41 & Age <61 ~ "50-60",
  TRUE ~ "0"))
  • Created additional variable, ‘Age_Bracket’ to simplify creation of graphs.
ggplot(data = employee) +
  geom_bar(mapping = aes(x = Age_Bracket, fill = Gender),position = "dodge") +
  scale_fill_manual(values = c("#D50057", "#003595")) + 
  theme(plot.title = element_text(hjust = 0.5)) +
  ggtitle("Age by Gender") +
  xlab("Age") +
  ylab("# of people") +
  theme_minimal() +
  theme(axis.line = element_line(colour = "grey50")) +
  theme_minimal() +
  theme(panel.grid.major = element_line(color = "white"),
        axis.title.x = element_text(colour = "#60269E", face = "bold"),
        axis.title.y = element_text(colour = "#60269E", face = "bold"),
        axis.text = element_text(colour = "#666666"),
        axis.ticks = element_line(colour = "#666666"),
        axis.line = element_line(colour = "#666666"),
        title = element_text(colour = "#60269E", face = "bold"),
        plot.title = element_text(face = "bold")) +
  theme(plot.title = element_text(hjust = 0.5))

  • As can be seen above, there is more males than females working at BioMarin Pharmaceuticals.

  • Highest population in bracket 30 - 40.

Mean Monthly Pay and Performance Rating sorted by Gender

Mean_Pay_By_Gender <- employee %>%
  group_by(Gender) %>%
  summarise(mean_monthly_pay = mean(MonthlyRate)) %>%
  arrange(mean_monthly_pay) 

Mean_Performance_By_Gender <- employee %>%
  group_by(Gender) %>%
  summarise(mean_performance_rating = round (mean(PerformanceRating))) %>%
  arrange(mean_performance_rating) 
  

Table_1_Gender <- merge(x = Mean_Pay_By_Gender, y = Mean_Performance_By_Gender, by ="Gender", all=TRUE) 


knitr::kable(Table_1_Gender, "pipe", col.names = c("Gender", "Mean Monthly Pay", "Performance Rating"), align = c("c", "c", "c"), caption = "Mean Pay and Performance by Gender")
Mean Pay and Performance by Gender
Gender Mean Monthly Pay Performance Rating
Female 14674.60 3
Male 14072.11 3
  • Women are on average paid more than men, despite the performance rating for both being 3.

Mean Monthly Pay and Performance Rating sorted by Age Bracket

Mean_Pay_By_Age <- employee %>%
  group_by(Age_Bracket) %>%
  summarise(mean_monthly_pay = mean(MonthlyRate)) %>%
  arrange(mean_monthly_pay) 

Mean_Performance_By_Age <- employee %>%
  group_by(Age_Bracket) %>%
  summarise(mean_performance = round(mean(PerformanceRating), 2)) %>%
  arrange(mean_performance)

Table_2_Age <- merge(x = Mean_Pay_By_Age, y = Mean_Performance_By_Age, by ="Age_Bracket", all=TRUE) 

knitr::kable(Table_2_Age, "pipe", col.names = c("Age Bracket", "Mean Monthly Pay", "Performance Rating"), align = c("c", "c", "c"), caption = "Mean Pay and Performance by Age Bracket")
Mean Pay and Performance by Age Bracket
Age Bracket Mean Monthly Pay Performance Rating
18-20 15470.54 3.14
20-30 14039.06 3.15
30-40 14273.64 3.15
40-50 13794.02 3.15
50-60 16112.22 3.17

  • Age bracket of 50 - 60 earns the most on average, with a 3.17 performance rating.

  • The second highest earning age bracket is 18 - 20, despite having the lowest performance rating (3.14).

  • Age bracket 40 - 50 earns the least on average.

Work Performance by Age Bracket

geomcol_annotation <- data.frame(
  x = c(3.68),
  y = c(5),
  label = c(" highest 
            performance")) 

ggplot(data = Mean_Performance_By_Age) +
  geom_col(mapping = aes(x = mean_performance, y = Age_Bracket, fill = Age_Bracket)) +
  geom_text(aes(x = mean_performance, y = Age_Bracket, 
                label = mean_performance), hjust = -0.10, color = "#666666") +
  xlim(0,4) +
  ggtitle("Work Performance by Age Bracket") +
  xlab("Mean Performance") +
  ylab("Age Bracket") +
  geom_text(data = geomcol_annotation, aes( x=x, y=y, label=label),                 , 
            color="#C017A2", 
            size = 3 , fontface = "bold") +
  theme_minimal() +
  theme(panel.grid.major = element_line(color = "white"),
        axis.title.x = element_text(colour = "#60269E", face = "bold"),
        axis.title.y = element_text(colour = "#60269E", face = "bold"),
        axis.text = element_text(colour = "#666666"),
        axis.ticks = element_line(colour = "#666666"),
        axis.line = element_line(colour = "#666666"),
        title = element_text(colour = "#60269E", face = "bold"),
        plot.title = element_text(face = "bold")) +
  theme(plot.title = element_text(hjust = 0.5)) +
  gghighlight(Age_Bracket == "50-60") +
  scale_fill_manual(values = c("#C017A2"))
Warning: Tried to calculate with group_by(), but the calculation failed.
Falling back to ungrouped filter operation...
Warning: Could not calculate the predicate for layer 3; ignored

Annual Pay by Department (Does Travel frequency Affect Pay?)

employee <- mutate(employee, YearlyRate = MonthlyRate*12)
  • Created variable ‘YearlyRate’ for the creation of ‘Annual Pay by Department’.
ggplot(data = employee) +
  geom_boxplot(mapping = aes(x = Department, y = YearlyRate, fill = BusinessTravel)) +
  scale_fill_manual(values = c("#FF4438", "#C017A2", "#0197f6")) + 
  ggtitle ("Annual Pay by Department") +
  xlab("Department") +
  ylab("Annual Pay, in '000") +
  scale_y_continuous(limits = c(0, 350000), breaks = c(0, 87500, 175000, 262500, 350000), labels = c('0','87.5', '175', '262.5', '350')) +
  theme_minimal() +
  theme(plot.title = element_text(hjust = 0.5)) + 
  theme(panel.grid.major = element_line(color = "white"),
        axis.title.x = element_text(colour = "#60269E", face = "bold"),
        axis.title.y = element_text(colour = "#60269E", face = "bold"),
        axis.text = element_text(colour = "#666666"),
        axis.ticks = element_line(colour = "#666666"),
        axis.line = element_line(colour = "#666666"),
        title = element_text(colour = "#60269E", face = "bold"),
        plot.title = element_text(face = "bold"))

  • The department of sales earns the most on average, with those travelling rarely earning the most.

  • The department of Human Resources earns the least on average, with those travelling frequently earning the least.

  • In most cases, travelling frequently lowers annual salary (across all departments).

Time at Company and Time Spent in Current Role

scatterplot_annotation <- data.frame(
  x = c(10),
  y = c(5),
  label = c("Positive correlation")) 

scatterplot <- ggplot(data = employee) + 
  geom_point (mapping = aes(x = YearsInCurrentRole, y = YearsAtCompany)) +
  ggtitle("Correlation between Time at Company and Time in Current Role (in years)") +
  theme(plot.title = element_text(hjust = 0.5)) + 
  xlab("Time in Current Role") +
  ylab("Time at Compnay") +
  xlim(0, 18) + 
  annotate("segment", x = 0, xend = 18, y = 0, yend = 20, colour = "#C017A2", linewidth = 3, alpha = 0.6, arrow = arrow()) + 
  geom_text(data = scatterplot_annotation, aes( x=x, y=y, label=label),                 , 
            color = "#C017A2", 
            size = 5 , fontface="bold" ) +
  theme_minimal() +
  theme(panel.grid.major = element_line(color = "white"),
        axis.title.x = element_text(colour = "#60269E", face = "bold"),
        axis.title.y = element_text(colour = "#60269E", face = "bold"),
        axis.text = element_text(colour = "#666666"),
        axis.ticks = element_line(colour = "#666666"),
        axis.line = element_line(colour = "#666666"),
        title = element_text(colour = "#60269E", face = "bold"),
        plot.title = element_text(face = "bold"))

scatterplot

  • There is a positive correlation between Years spent in Current Role and Overall Years Spent in the Company.

  • This suggests that employees working in the same role for multiple years are more likely to keep working at BioMarin Pharmaceuticals for the long run.

Connection Between Job Satisfaction and Training Level

  • Created variable ‘Training_Level’ to simplify graph creation.
summarize(employee, min(TrainingTimesLastYear)) #0#
# A tibble: 1 × 1
  `min(TrainingTimesLastYear)`
                         <dbl>
1                            0
summarize(employee, max(TrainingTimesLastYear)) #6#
# A tibble: 1 × 1
  `max(TrainingTimesLastYear)`
                         <dbl>
1                            6
employee <- mutate(employee, Training_Level = case_when(
  TrainingTimesLastYear = 0 &  TrainingTimesLastYear <3 ~ "Low",
  TrainingTimesLastYear = 2 &  TrainingTimesLastYear <5 ~ "Medium",
  TrainingTimesLastYear = 4 &  TrainingTimesLastYear <7 ~ "High",
  TRUE ~ "0")) #training level for easier histogram creation#
ggplot(data = employee) +
  geom_histogram(mapping = aes(x = JobSatisfaction, fill = Training_Level), binwidth = 1, colour = "black") +
  scale_x_continuous(limits = c(0, 5), breaks = c(0, 1, 2, 3, 4)) +
  scale_fill_manual(values = c("#60269E", "#C017A2")) +
  ggtitle("Correlation between Job Satisfaction and Training") +
  theme(plot.title = element_text(hjust = 0.5)) + 
  xlab("Job Satisfcation") +
  ylab("Count of Training") +
  theme_minimal() +
  theme(panel.grid.major = element_line(color = "white"),
        axis.title.x = element_text(colour = "#60269E", face = "bold"),
        axis.title.y = element_text(colour = "#60269E", face = "bold"),
        axis.text = element_text(colour = "#666666"),
        axis.ticks = element_line(colour = "#666666"),
        axis.line = element_line(colour = "#666666"),
        title = element_text(colour = "#60269E", face = "bold"),
        plot.title = element_text(face = "bold"))
Warning: Removed 4 rows containing missing values or values outside the scale range
(`geom_bar()`).

  • Employees with higher Job Satisfaction received more training than employees with lower Job Satisfaction.

Connection Between Performance Rating and Training Level

ggplot(data = employee) +
  geom_histogram(mapping = aes(x = PerformanceRating, fill = Training_Level), binwidth = 1, colour = "black") +
  scale_x_continuous(limits = c(0, 5), breaks = c(0, 1, 2, 3, 4)) +
  scale_fill_manual(values = c("#60269E", "#C017A2")) +
  ggtitle("Correlation between Job Performance and Training") +
  theme(plot.title = element_text(hjust = 0.5)) + 
  xlab("Performance Rating") +
  ylab("Count of Training") +
  theme_minimal() +
  theme(panel.grid.major = element_line(color = "white"),
        axis.title.x = element_text(colour = "#60269E", face = "bold"),
        axis.title.y = element_text(colour = "#60269E", face = "bold"),
        axis.text = element_text(colour = "#666666"),
        axis.ticks = element_line(colour = "#666666"),
        axis.line = element_line(colour = "#666666"),
        title = element_text(colour = "#60269E", face = "bold"),
        plot.title = element_text(face = "bold"))
Warning: Removed 4 rows containing missing values or values outside the scale range
(`geom_bar()`).

  • Those who received a higher Level of Training didn’t achieve a higher Performance Rating.
  • Hence, higher Training Levels don’t result in Higher Performance Ratings.

Attrition in each Job Role

ggplot(data = employee) +
  geom_jitter(mapping = aes(x = JobRole, y = JobSatisfaction, fill = Attrition, colour = Attrition)) +
  scale_color_manual(values = c("#D2E0ff", "#C017A2")) +
  theme_minimal() +
  ggtitle("Attrition by Job Role") +
  theme(plot.title = element_text(hjust = 0.5)) + #title in center#
  xlab("Job Role") +
  ylab("Job Satisfaction") +
  theme(axis.text.x=element_text(angle=-90, vjust = 0.5, hjust = 0),
        panel.grid.major = element_line(color = "white"),
        axis.title.x = element_text(colour = "#60269E", face = "bold"),
        axis.title.y = element_text(colour = "#60269E", face = "bold"),
        axis.text = element_text(colour = "#666666"),
        axis.ticks = element_line(colour = "#666666"),
        axis.line = element_line(colour = "#666666"),
        title = element_text(colour = "#60269E", face = "bold"),
        plot.title = element_text(face = "bold"))

  • Laboratory Technicians, Research Scientists, Sales Executives and Sales Representatives have the highest Attrition Levels.

  • Research Directors, Managers, Healthcare Representatives and Manufacturing Directors have the lowest Attrition Levels.

Table_3 <- employee %>%
  group_by(JobRole) %>%
  summarise(mean_job_sat = round (mean(JobSatisfaction),2)) %>%
  arrange(mean_job_sat)

knitr::kable(Table_3, "pipe", col.names = c("Job Role", "Job Satisfaction"), align = c("l", "c"), caption = "Job Satisfaction by Job Role (lowest to highest)")
Job Satisfaction by Job Role (lowest to highest)
Job Role Job Satisfaction
Human Resources 2.56
Manufacturing Director 2.68
Laboratory Technician 2.69
Research Director 2.70
Manager 2.71
Sales Representative 2.73
Sales Executive 2.75
Research Scientist 2.77
Healthcare Representative 2.79

‘Laboratory Technicians, Research Scientists, Sales Executives and Sales Representatives have the highest Attrition Levels.’

  • Based on highest Attrition Levels, Research Scientists, Sales Executives and Sales Representatives are scored the highest in terms of Job Satisfaction, whilst Laboratory Technicians have 3rd lowest Job Satisfaction Ratings.

‘Research Directors, Managers, Healthcare Representatives and Manufacturing Directors have the lowest Attrition Levels.’

  • Based on lowest Attrition Levels, Manufacturing Directors, Research directors and Managers have increasing Job Satisfaction (named in order). healthcare Representatives have the highest Job Satisfaction of all Job Roles.