Sleep Health & Lifestyle – Exploratory Analysis

Author

Pascal Hermann Kouogang Tafo

INTRODUCTION

This assignment analyses the Sleep Health & Lifestyle Dataset to understand how lifestyle factors such as occupational role and psychological stress could shape two core sleep outcomes: sleep duration and self-rated sleep quality.Understanding these relationships is relevant for us as individual and corporations when creating workplace wellness programs to improve employees performance.

APPROACH

To conduct the analysis of the dataset, i will implement a structured data science pipeline using the “tidyverse” framework as followed:

Load the CSV dataset in R and commit to an existing GitHub repository ensuring its accessibility at anytime .
Rename some variables by removing “_” and follow a consistent naming convention.
Transform my dataset from Wide to Long format using the pivot_longer function to convert the sleep duration and quality into a single metric column for easier modelling and faceted plotting.
Analyze and Visualize the Correlation between Sleep Duration and Quality related to Occupation. Interpret the result
Analyze and Visualize the Correlation between Sleep Duration and Quality related to Stress Level. Interpret the result

Load Library

library(tidyverse)

Warning: package 'tidyverse' was built under R version 4.5.2

Warning: package 'tibble' was built under R version 4.5.2

Warning: package 'tidyr' was built under R version 4.5.2

Warning: package 'readr' was built under R version 4.5.2

Warning: package 'purrr' was built under R version 4.5.2

Warning: package 'dplyr' was built under R version 4.5.2

Warning: package 'stringr' was built under R version 4.5.2

Warning: package 'forcats' was built under R version 4.5.2

Warning: package 'lubridate' was built under R version 4.5.2

── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
✔ dplyr     1.2.0     ✔ readr     2.2.0
✔ forcats   1.0.1     ✔ stringr   1.6.0
✔ ggplot2   4.0.2     ✔ tibble    3.3.1
✔ lubridate 1.9.5     ✔ tidyr     1.3.2
✔ purrr     1.2.1     
── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
✖ dplyr::filter() masks stats::filter()
✖ dplyr::lag()    masks stats::lag()
ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors

library(janitor) # for cleaning column names

Warning: package 'janitor' was built under R version 4.5.2


Attaching package: 'janitor'

The following objects are masked from 'package:stats':

    chisq.test, fisher.test

Load and clean the data

I will read the file from my GitHub. For the cleaning i will need the help of the LLM Claude because some columns names are inconsistent and some variables are compounds. Here is the prompt : “Clean this dataset by renaming to snake_case and parsing compounds variables.”

# Read file

url <- "https://raw.githubusercontent.com/Pascaltafo2025/PROJECT-2--TIDY-DATA-ANALYSIS/refs/heads/main/Sleep_health_and_lifestyle_dataset.csv"

Sleep_health_and_lifestyle <- read_csv(url)

Rows: 374 Columns: 13
── Column specification ────────────────────────────────────────────────────────
Delimiter: ","
chr (5): Gender, Occupation, BMI Category, Blood Pressure, Sleep Disorder
dbl (8): Person ID, Age, Sleep Duration, Quality of Sleep, Physical Activity...

ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

head(Sleep_health_and_lifestyle,10)

# A tibble: 10 × 13
   `Person ID` Gender   Age Occupation       `Sleep Duration` `Quality of Sleep`
         <dbl> <chr>  <dbl> <chr>                       <dbl>              <dbl>
 1           1 Male      27 Software Engine…              6.1                  6
 2           2 Male      28 Doctor                        6.2                  6
 3           3 Male      28 Doctor                        6.2                  6
 4           4 Male      28 Sales Represent…              5.9                  4
 5           5 Male      28 Sales Represent…              5.9                  4
 6           6 Male      28 Software Engine…              5.9                  4
 7           7 Male      29 Teacher                       6.3                  6
 8           8 Male      29 Doctor                        7.8                  7
 9           9 Male      29 Doctor                        7.8                  7
10          10 Male      29 Doctor                        7.8                  7
# ℹ 7 more variables: `Physical Activity Level` <dbl>, `Stress Level` <dbl>,
#   `BMI Category` <chr>, `Blood Pressure` <chr>, `Heart Rate` <dbl>,
#   `Daily Steps` <dbl>, `Sleep Disorder` <chr>

## Rename to snake_case to insure consistency

Sleep_health_and_lifestyle <- Sleep_health_and_lifestyle |>
  rename(
    person_id             = `Person ID`,
    gender                = Gender,
    age                   = Age,
    occupation            = Occupation,
    sleep_duration_hrs    = `Sleep Duration`,
    sleep_quality_score   = `Quality of Sleep`,
    physical_activity_min = `Physical Activity Level`,
    stress_level          = `Stress Level`,
    bmi_category          = `BMI Category`,
    blood_pressure        = `Blood Pressure`,
    heart_rate_bpm        = `Heart Rate`,
    daily_steps           = `Daily Steps`,
    sleep_disorder        = `Sleep Disorder`
  )

# Clean 
Sleep_health_and_lifestyle_clean <- Sleep_health_and_lifestyle |>
  mutate(
    # Decision 1: harmonise synonymous BMI labels
    
    bmi_category  = recode(bmi_category, "Normal" = "Normal Weight"),
    bmi_category  = factor(bmi_category,
                           levels = c("Normal Weight", "Overweight", "Obese")),
    # Decision 2: NA sleep_disorder → explicit "None"
    sleep_disorder = replace_na(sleep_disorder, "None"),
    sleep_disorder = factor(sleep_disorder,
                            levels = c("None", "Insomnia", "Sleep Apnea")),
    # Decision 3: parse compound blood pressure string
    bp_systolic   = as.integer(str_split_fixed(blood_pressure, "/", 2)[, 1]),
    bp_diastolic  = as.integer(str_split_fixed(blood_pressure, "/", 2)[, 2]),
    # Decision 4: type coercions
    gender        = factor(gender),
    occupation    = factor(occupation),
    stress_level  = as.integer(stress_level),
    person_id     = as.character(person_id)
  ) |>
  select(-blood_pressure)

head(Sleep_health_and_lifestyle_clean,10)

# A tibble: 10 × 14
   person_id gender   age occupation      sleep_duration_hrs sleep_quality_score
   <chr>     <fct>  <dbl> <fct>                        <dbl>               <dbl>
 1 1         Male      27 Software Engin…                6.1                   6
 2 2         Male      28 Doctor                         6.2                   6
 3 3         Male      28 Doctor                         6.2                   6
 4 4         Male      28 Sales Represen…                5.9                   4
 5 5         Male      28 Sales Represen…                5.9                   4
 6 6         Male      28 Software Engin…                5.9                   4
 7 7         Male      29 Teacher                        6.3                   6
 8 8         Male      29 Doctor                         7.8                   7
 9 9         Male      29 Doctor                         7.8                   7
10 10        Male      29 Doctor                         7.8                   7
# ℹ 8 more variables: physical_activity_min <dbl>, stress_level <int>,
#   bmi_category <fct>, heart_rate_bpm <dbl>, daily_steps <dbl>,
#   sleep_disorder <fct>, bp_systolic <int>, bp_diastolic <int>

Transform my dataset from Wide to Long format using the pivot_longer function

Sleep_health_and_lifestyle_clean_long <- Sleep_health_and_lifestyle_clean |>
  pivot_longer(
    cols      = c(sleep_duration_hrs, sleep_quality_score),
    names_to  = "sleep_metric",
    values_to = "metric_value"
  ) |>
  mutate(
    sleep_metric = recode(sleep_metric,
      "sleep_duration_hrs"  = "Sleep Duration (hrs)",
      "sleep_quality_score" = "Sleep Quality (score)"
    )
  )

head(Sleep_health_and_lifestyle_clean_long,10)

# A tibble: 10 × 14
   person_id gender   age occupation          physical_activity_min stress_level
   <chr>     <fct>  <dbl> <fct>                               <dbl>        <int>
 1 1         Male      27 Software Engineer                      42            6
 2 1         Male      27 Software Engineer                      42            6
 3 2         Male      28 Doctor                                 60            8
 4 2         Male      28 Doctor                                 60            8
 5 3         Male      28 Doctor                                 60            8
 6 3         Male      28 Doctor                                 60            8
 7 4         Male      28 Sales Representati…                    30            8
 8 4         Male      28 Sales Representati…                    30            8
 9 5         Male      28 Sales Representati…                    30            8
10 5         Male      28 Sales Representati…                    30            8
# ℹ 8 more variables: bmi_category <fct>, heart_rate_bpm <dbl>,
#   daily_steps <dbl>, sleep_disorder <fct>, bp_systolic <int>,
#   bp_diastolic <int>, sleep_metric <chr>, metric_value <dbl>

Analyze and Visualize the Correlation between Sleep Duration and Quality related to Occupation

# Compute sleep statistical summary by occupation

occupation_summary <- Sleep_health_and_lifestyle_clean %>%
  group_by(occupation) %>%
  summarise(
    n                   = n(),
    mean_sleep_duration = mean(sleep_duration_hrs,  na.rm = TRUE),
    mean_sleep_quality  = mean(sleep_quality_score, na.rm = TRUE),
    mean_stress         = mean(stress_level,         na.rm = TRUE),
    .groups = "drop"
  )
occupation_summary

# A tibble: 11 × 5
   occupation               n mean_sleep_duration mean_sleep_quality mean_stress
   <fct>                <int>               <dbl>              <dbl>       <dbl>
 1 Accountant              37                7.11               7.89        4.59
 2 Doctor                  71                6.97               6.65        6.73
 3 Engineer                63                7.99               8.41        3.89
 4 Lawyer                  47                7.41               7.89        5.06
 5 Manager                  1                6.9                7           5   
 6 Nurse                   73                7.06               7.37        5.55
 7 Sales Representative     2                5.9                4           8   
 8 Salesperson             32                6.40               6           7   
 9 Scientist                4                6                  5           7   
10 Software Engineer        4                6.75               6.5         6   
11 Teacher                 40                6.69               6.98        4.53

Stress-level averages

stress_summary <- Sleep_health_and_lifestyle_clean %>%
  group_by(stress_level) %>%
  summarise(
    n                   = n(),
    mean_sleep_duration = mean(sleep_duration_hrs,  na.rm = TRUE),
    mean_sleep_quality  = mean(sleep_quality_score, na.rm = TRUE),
    .groups = "drop"
  )

stress_summary

# A tibble: 6 × 4
  stress_level     n mean_sleep_duration mean_sleep_quality
         <int> <int>               <dbl>              <dbl>
1            3    71                8.23               8.97
2            4    70                7.03               7.67
3            5    67                7.48               7.90
4            6    46                7.45               7   
5            7    50                6.47               6   
6            8    70                6.05               5.86

Visualization

# Overall Pearson correlation
overall_r <- cor.test(Sleep_health_and_lifestyle_clean$sleep_duration_hrs, Sleep_health_and_lifestyle_clean$sleep_quality_score)
cat(sprintf("Overall Pearson r = %.3f  (p = %.2e)\n\n",
            overall_r$estimate, overall_r$p.value))

Overall Pearson r = 0.883  (p = 2.17e-124)

# =============================================================================
# STEP 7 – Visualisation: Sleep Metrics ~ Occupation
# =============================================================================

# ── OCCUPATION PLOT A: raw scatter coloured by occupation ────────────────────
occ_palette <- c(
  "Accountant"          = "#4C72B0",
  "Doctor"              = "#DD8452",
  "Engineer"            = "#55A868",
  "Lawyer"              = "#C44E52",
  "Manager"             = "#8172B2",
  "Nurse"               = "#937860",
  "Sales Representative"= "#DA8BC3",
  "Salesperson"         = "#E7BA52",
  "Scientist"           = "#8C8C8C",
  "Software Engineer"   = "#CCB974",
  "Teacher"             = "#64B5CD"
)

ggplot(Sleep_health_and_lifestyle_clean,
  aes(x = sleep_duration_hrs,
      y = sleep_quality_score,
      colour = occupation)) +
  geom_point(alpha = 0.55, size = 2.2) +
  geom_smooth(data   = Sleep_health_and_lifestyle_clean,
              aes(x  = sleep_duration_hrs,
                  y  = sleep_quality_score),
              method = "lm", formula = y ~ x,
              colour = "black", linewidth = 0.9,
              linetype = "dashed", se = TRUE, inherit.aes = FALSE) +
  scale_colour_manual(values = occ_palette, name = "Occupation") +
  scale_x_continuous(breaks = seq(5, 9, 0.5)) +
  scale_y_continuous(breaks = 1:10) +
  labs(
    title    = "Sleep Duration vs Quality of Sleep by Occupation",
    subtitle = sprintf("Individual observations  |  Overall Pearson r = %.2f",
                       overall_r$estimate),
    x        = "Sleep Duration (hours per night)",
    y        = "Quality of Sleep (1 = Poor -> 10 = Excellent)",
    caption  = "Source: Sleep Health & Lifestyle Dataset"
  ) +
  theme_minimal(base_size = 12) +
  theme(
    plot.title      = element_text(face = "bold", size = 14),
    plot.subtitle   = element_text(colour = "grey40"),
    legend.position = "right"
  )

# ── OCCUPATION PLOT B: grouped bar chart – mean duration & quality ────────────
occ_long <- occupation_summary %>%
  pivot_longer(cols = c(mean_sleep_duration, mean_sleep_quality),
               names_to  = "metric",
               values_to = "mean_value") %>%
  mutate(
    metric = recode(metric,
      "mean_sleep_duration" = "Avg Sleep Duration (hrs)",
      "mean_sleep_quality"  = "Avg Sleep Quality (score)"
    ),
    occupation = fct_reorder(occupation, mean_value, .fun = max)
  )
 ggplot(occ_long,
  aes(x    = mean_value,
      y    = occupation,
      fill = metric)) +
  geom_col(position = "dodge", width = 0.65, alpha = 0.88) +
  scale_fill_manual(values = c("Avg Sleep Duration (hrs)"  = "#4C72B0",
                               "Avg Sleep Quality (score)" = "#DD8452"),
                    name   = "Sleep Metric") +
  scale_x_continuous(limits = c(0, 10), breaks = seq(0, 10, 2)) +
  labs(
    title   = "Mean Sleep Duration & Quality by Occupation",
    x       = "Mean Value",
    y       = NULL,
    caption = "Source: Sleep Health & Lifestyle Dataset"
  ) +
  theme_minimal(base_size = 12) +
  theme(
    plot.title      = element_text(face = "bold", size = 14),
    legend.position = "bottom"
  )