DATA_608_Story_04

Author

Henock Montcho

Published

May 22, 2026

Code

# Load libraries
library(tidytext)
library(readxl)
library(tidyverse)
library(sf)
library(tigris)
library(viridis)
library(usmap) 
library(ggplot2)
library(forcats)
library(scales)
library(dplyr)

setwd("C:/Users/month/Downloads")

Data_Scientist <- read_excel("C:/Users/month/Downloads/Data_Scientist.xlsx")
Data_Scientist$Salary <- as.numeric(gsub("[$,\\s]", "", Data_Scientist$Salary))
head(Data_Scientist)

# A tibble: 6 × 2
  State      Salary
  <chr>       <dbl>
1 Alabama    111249
2 Alaska     132182
3 Arizona    114378
4 Arkansas   101493
5 California 121131
6 Colorado   129062

Code

Data_Engineer <- read_excel("C:/Users/month/Downloads/Data_Engineer.xlsx")
Data_Engineer$Salary <- as.numeric(gsub("[$,\\s]", "", Data_Engineer$Salary))
head(Data_Engineer)

# A tibble: 6 × 2
  State      Salary
  <chr>       <dbl>
1 Alabama    117573
2 Alaska     139697
3 Arizona    120881
4 Arkansas   107263
5 California 128018
6 Colorado   136399

Code

Data_Analyst <- read_excel("C:/Users/month/Downloads/Data_Analyst.xlsx")
Data_Analyst$Salary <- as.numeric(gsub("[$,\\s]", "", Data_Analyst$Salary))
head(Data_Analyst)

# A tibble: 6 × 2
  State      Salary
  <chr>       <dbl>
1 Alabama     74904
2 Alaska      88999
3 Arizona     77011
4 Arkansas    68335
5 California  81558
6 Colorado    86898

Code

Business_Analyst <- read_excel("C:/Users/month/Downloads/Business_Analyst.xlsx")
Business_Analyst$Salary <- as.numeric(gsub("[$,\\s]", "", Business_Analyst$Salary))
head(Business_Analyst)

# A tibble: 6 × 2
  State      Salary
  <chr>       <dbl>
1 Alabama     89426
2 Alaska     106253
3 Arizona     91942
4 Arkansas    81584
5 California  97370
6 Colorado   103744

Code

Data_Practitioner <- Data_Scientist |>
  left_join(Data_Engineer, by = "State") |>
  left_join(Data_Analyst, by = "State") |>
  left_join(Business_Analyst, by = "State") |>
  rename(Data_Scientist   = Salary.x,
    Data_Engineer    = Salary.y,
    Data_Analyst     = Salary.x.x,
    Business_Analyst = Salary.y.y)

# Convert to long

Data_Practitioner_Salaries <- Data_Practitioner |>
  pivot_longer(
    cols = c(Data_Scientist, Data_Engineer, Data_Analyst, Business_Analyst),
    names_to = "role",
    values_to = "salary") |>
  mutate(
    state = str_trim(as.character(State)),
    salary = as.numeric(salary)
  )

head(Data_Practitioner_Salaries)

# A tibble: 6 × 4
  State   role             salary state  
  <chr>   <chr>             <dbl> <chr>  
1 Alabama Data_Scientist   111249 Alabama
2 Alabama Data_Engineer    117573 Alabama
3 Alabama Data_Analyst      74904 Alabama
4 Alabama Business_Analyst  89426 Alabama
5 Alaska  Data_Scientist   132182 Alaska 
6 Alaska  Data_Engineer    139697 Alaska

Code

# Roles Boxplot 

palette <- c("#9BB8AD", "#A39FA1", "#DEB3A0", "#FEC6AF")

bp_role <- ggplot(Data_Practitioner_Salaries, aes(x=" ", y = salary, group = role)) + 
  geom_boxplot(aes(fill = role)) + 
  theme_minimal() +
  scale_y_continuous(labels = label_comma()) +
  facet_grid(. ~ role) +
  scale_fill_manual(values=palette) +
  theme(legend.position = "none") +
  theme(text = element_text(size=12), axis.title=element_text(size=12))
bp_role

Comment:

Overall and nationally, Data Engineers and Data Scientists earn meaningfully higher and more variable salaries than Business Analysts and Data Analysts. The distribution shapes also show that technical depth and engineering complexity correlate with both higher medians and wider pay ranges.
- BA roles tend to be more structured and less specialized, so compensation is predictable and capped.
- DA roles can scale upward when analytics blends into engineering or modeling, but the core market rate remains modest.
- DE is one of the most in‑demand roles; salary reflects the complexity of building scalable data systems.
- DS compensation varies widely depending on whether the role is analytics‑focused or ML‑engineering‑focused, but the ceiling is high.

Code

Data_Scientist1 <- select(Data_Practitioner, c("Data_Scientist", "State"))
Data_Engineer1  <- select(Data_Practitioner, c("Data_Engineer", "State"))
Data_Analyst1   <- select(Data_Practitioner, c("Data_Analyst", "State"))
Business_Analyst1 <- select(Data_Practitioner, c("Business_Analyst", "State"))


# Bar chart Data Scientist

bc_DS <- ggplot(Data_Scientist1, aes(x = fct_reorder(Data_Scientist1$State, Data_Scientist1$Data_Scientist), y = Data_Scientist1$Data_Scientist, fill = State)) +
  geom_col() +                         
  coord_flip() +
  scale_y_continuous(labels = label_comma()) +
  theme_minimal() +
  theme(
    legend.position = "none",
    text = element_text(size = 8),
    axis.title = element_text(size = 12),
    plot.title = element_text(size = 10)
  ) +
  labs(
    title = "US Average Data Scientist Salaries by State / Territory",
    x = "State or Territory",
    y = "Annual Average Salary"
  )

bc_DS

Code

# Bar Chart Data Engineer

bc_DE <- ggplot(Data_Engineer1, aes(x = fct_reorder(Data_Engineer1$State, Data_Engineer1$Data_Engineer), y = Data_Engineer1$Data_Engineer, fill = State)) +
  geom_col() +                         
  coord_flip() +
  scale_y_continuous(labels = label_comma()) +
  theme_minimal() +
  theme(
    legend.position = "none",
    text = element_text(size = 8),
    axis.title = element_text(size = 12),
    plot.title = element_text(size = 10)
  ) +
  labs(
    title = "US Average Data Engineer Salaries by State / Territory",
    x = "State or Territory",
    y = "Annual Average Salary"
  )

bc_DE

Code

# Bar Chart Data analyst
bc_DA <- ggplot(Data_Analyst1, aes(x = fct_reorder(Data_Analyst1$State, Data_Analyst1$Data_Analyst), y = Data_Analyst, fill = State)) +
  geom_col() +                         
  coord_flip() +
  scale_y_continuous(labels = label_comma()) +
  theme_minimal() +
  theme(
    legend.position = "none",
    text = element_text(size = 8),
    axis.title = element_text(size = 12),
    plot.title = element_text(size = 10)
  ) +
  labs(
    title = "US Average Data Analyst Salaries by State / Territory",
    x = "State or Territory",
    y = "Annual Average Salary"
  )

bc_DA

Code

# Bar Chart Business Analyst
bc_BA <- ggplot(Business_Analyst1, aes(x = fct_reorder(Business_Analyst1$State, Business_Analyst1$Business_Analyst), y = Business_Analyst, fill = State)) +
  geom_col() +                         
  coord_flip() +
  scale_y_continuous(labels = label_comma()) +
  theme_minimal() +
  theme(
    legend.position = "none",
    text = element_text(size = 8),
    axis.title = element_text(size = 12),
    plot.title = element_text(size = 10)
  ) +
  labs(
    title = "US Average Business Analyst Salaries by State / Territory",
    x = "State or Territory",
    y = "Annual Average Salary"
  )

bc_BA

Code

### Bar Chart for the average of the overall Data Practitioner Salaries by States.

Data_Practitioner$Avg_Salary <- rowMeans(
  Data_Practitioner[, c("Data_Scientist", "Data_Engineer", "Data_Analyst", "Business_Analyst")],
  na.rm = TRUE
)

bc_state <- ggplot(Data_Practitioner, 
                   aes(x = fct_reorder(State, Avg_Salary), 
                       y = Avg_Salary, 
                       fill = State)) +
  geom_col() +                         
  coord_flip() +
  scale_y_continuous(labels = label_comma()) +
  theme_minimal() +
  theme(
    legend.position = "none",
    text = element_text(size = 8),
    axis.title = element_text(size = 12),
    plot.title = element_text(size = 10)
  ) +
  labs(
    title = "US Average Data Practitioner Salaries by State / Territory",
    x = "State or Territory",
    y = "Average Salary Across Roles"
  )

bc_state

Code

# Faceted bar chart of Top 20 States of all four roles in one plot (ranked by Data Scientist Role)

top20_states <- Data_Practitioner  |>
  arrange(desc(Data_Scientist))  |>
  slice(1:20)  |>
  pull(State)

dp_top20 <- Data_Practitioner  |>
  filter(State %in% top20_states)  |>
  pivot_longer(
    cols = c(Data_Scientist, Data_Engineer, Data_Analyst, Business_Analyst),
    names_to = "Role",
    values_to = "Salary"
  )

ggplot(dp_top20, aes(x = fct_reorder(State, Salary), y = Salary, fill = State)) +
  geom_col() +
  coord_flip() +
  facet_wrap(~ Role, scales = "free_y") +
  scale_y_continuous(labels = scales::label_comma()) +
  theme_minimal() +
  theme(legend.position = "none") +
  labs(
    title = "Top 20 Highest-Paying States for Data Practitioner Roles",
    x = "State",
    y = "Average Salary"
  )

Interpretation:
Geography amplifies salary differences but does not change the ranking of roles. High‑cost states lift all roles upward; low‑cost states compress salaries downward.

Code

dp_long <- Data_Practitioner  |>
  pivot_longer(
    cols = c(Data_Scientist, Data_Engineer, Data_Analyst, Business_Analyst),
    names_to = "Role",
    values_to = "Salary"
  )

  # Faceted bar chart

ggplot(dp_long, aes(x = Role, y = State, fill = Salary)) +
  geom_tile(color = "white") +
  scale_fill_viridis_c(option = "plasma") +
  theme_minimal() +
  labs(
    title = "Heatmap of Data Practitioner Salaries by State and Role",
    x = "Role",
    y = "State",
    fill = "Salary"
  )

Interpretation:
The heatmap confirms a stable national hierarchy of compensation across data roles, suggesting that role complexity and technical depth drive salary more than geography. Some states show uniformly higher salaries across all roles, indicated by lighter colors:

Likely high‑cost, high‑demand states (e.g., CA, NY, MA, WA).
These states show bright yellow for DS/DE and lighter greens for BA/DA.

Other states show uniformly lower salaries, indicated by darker purples:

Typically lower‑cost regions (e.g., southern or midwestern states).
Even Data Scientist salaries in these states fall into mid‑range colors.

Comment: The heatmap shows consistent salary stratification by role across all states—Data Scientists and Data Engineers earn the most, while Business Analysts and Data Analysts earn less, with noticeable geographic clusters where all roles command higher pay.

The heatmap quickly communicates relative differences across both roles and states.

Some states show greater spread between roles (large color contrast), while others show compressed ranges (similar tones across roles).
- High‑tech hubs show wider spreads; DS/DE salaries spike upward more dramatically.
- Smaller or lower‑demand states show narrow spreads, meaning the market differentiates less between roles.

Across nearly every state:

Data Scientist salaries are the highest band (yellow/bright tones).
Data Engineer salaries closely follow, often slightly lower but still in the upper range.
Data Analyst salaries fall into mid‑range colors (greens/purples).
Business Analyst salaries are consistently the lowest (darker purples).

Code

top20_states <- Data_Scientist1 %>%
  arrange(desc(Data_Scientist)) %>%
  slice(1:20) %>%
  pull(State)

Data_Scientist1 <- Data_Scientist1 %>%
  mutate(Group = ifelse(State %in% top20_states, "Top 20", "Other States"))

bc_DS <- ggplot(Data_Scientist1, 
                aes(x = fct_reorder(State, Data_Scientist), 
                    y = Data_Scientist, 
                    fill = Group)) +
  geom_col() +                         
  coord_flip() +
  scale_fill_manual(values = c("Top 20" = "#1f78b4",   # blue
                               "Other States" = "#b2df8a")) +  # green
  scale_y_continuous(labels = label_comma()) +
  theme_minimal() +
  theme(
    legend.position = "none",
    text = element_text(size = 8),
    axis.title = element_text(size = 12),
    plot.title = element_text(size = 10)
  ) +
  labs(
    title = "US Average Data Scientist Salaries by State / Territory",
    x = "State or Territory",
    y = "Annual Average Salary"
  )

bc_DS