1.1 Motivation

Our group focused on Tunisia, Egypt, Libya, and Sudan out of the six North African countries because these four nations show the most dynamic and contrasting transformations in how income factors relate to health outcomes. They capture diverse aspects of economic growth, inequality levels, and demographic transitions that make them ideal for analyzing how wealth distribution and income growth influence health indicators such as life expectancy and fertility rates. By comparing these four, we can uncover not only general regional patterns but also the unique national stories behind North Africa’s socio-economic evolution.

1.2 Data Description

The focus is on income categories and examining how financial inputs result in health outcomes across all four countries over time. The features used from the dataset include:

• Time: From 1800 to 2040. Reflects historical progress and key developmental milestones.
• Country: Four chosen North African countries — Egypt, Tunisia, Libya, and Sudan.
• income_level: Indicates whether a country is considered level 1–level 5.
• household_income: The mean income of all households in the country, usually expressed in local currency or USD. Indicates general wealth and living standards.
• gini: A measure of income inequality in a country.
  • 0 = perfect equality (everyone has the same income)
  • 100 = perfect inequality (one person has all the income)
• life_expectancy_at_birth: Average expected lifespan of a newborn.
• babies_per_woman: The average number of children a woman will have in her lifetime.
• Share of people on level 1–5: Percentage of the population whose income falls within each income level, from the poorest (Level 1) to the richest (Level 5).

1.3 Key Questions for Analysis

1.3.1 How did income_level, gini, babies_per_woman and life_expectancy_at_birth evolve from the 1800s to the 2000s across Tunisia, Egypt, Libya, and Sudan?
Observation: From the 1800s to the 2000s, all four North African countries transitioned from low-income, high-fertility societies with short life expectancy to more urbanized and economically developed nations with improved health outcomes. Between 1800 and 1950, income levels remained low, inequality moderate, and life expectancy rarely exceeded 40 years. After 1950, however, rapid socio-economic change began.

1.3.2 How do income-related factors influence babies_per_woman and life_expectancy_at_birth in North African countries over time?
Observation: Income plays a central role in shaping health outcomes across North African countries. As income levels and household earnings rise, people gain better access to healthcare, education, and nutrition - leading to longer life expectancy and lower fertility rates. Overall, the data reveals that sustainable health progress depends on both economic growth and social equality.

1.3.3 Do all four countries follow the same general trend between income inputs and health outcomes?
Observation: No, they don’t. While the overall relationship between income and health outcomes is generally positive across the four countries, the specific patterns differ notably due to each nation’s unique historical and socio-economic context. All countries show that higher income levels are associated with longer life expectancy and lower fertility rates, but their paths are not identical.

2.1 Libraries

To process and visualize the datasets effectively, we used several R libraries:

• readr: Efficiently read CSV datasets.
  
• dplyr: Manipulate and transform data, including filtering, joining, grouping, and summarizing variables.
  
• tidyr: Reshape and organize data for analysis and visualization.
  
• ggplot2: Create detailed and flexible plots.
  
• plotly: Add interactivity to visualizations for dynamic exploration.
  
• patchwork: Combine multiple plots into cohesive layouts.

library(dplyr)
library(readr)
library(ggplot2)
library(tidyr)
library(plotly)
library(patchwork)

2.2 Dataset Reading and Country Selection

We began by importing the datasets income_level.csv and world_data.csv. The analysis focused on four North African countries: Egypt, Libya, Sudan, and Tunisia. Using dplyr, we filtered both datasets to retain only the observations for these countries.

# Read datasets
income_level <- read_csv("income_level.csv", show_col_types = FALSE)
world_data <- read_csv("world_data.csv", show_col_types = FALSE)

# Define countries of interest
countries <- c("Egypt", "Libya", "Sudan", "Tunisia")

# Filter datasets
income_filtered <- income_level %>%
  filter(name %in% countries)

world_filtered <- world_data %>%
  filter(country %in% countries)

2.3 Combining and Verifying the Data

Since income_level only extends to 2040 while world_data continues to 2100, we aligned the datasets for the period 1800–2040 using a left join. This preserved all rows from income_level while appending corresponding variables from world_data.

combined_data_beforeclean <- income_filtered %>%
  left_join(world_filtered, by = c("name" = "country", "year" = "year"))

We verified the merge with several checks. These steps ensured that all entries were correctly aligned by country and year.

# Check unique country names
unique(income_filtered$name)
unique(world_filtered$country)

# Check year ranges
range(income_filtered$year)
range(world_filtered$year)

# Check missing values
sum(is.na(combined_data_beforeclean))
colSums(is.na(combined_data_beforeclean))
combined_data_beforeclean %>% filter(if_any(everything(), is.na))

# Spot-check specific rows
combined_data_beforeclean %>% 
  filter(name == "Egypt" & year %in% c(1800, 1900, 2000))

# Check for duplicate country-year pairs
combined_data_beforeclean %>%
  group_by(name, year) %>%
  summarise(n = n()) %>%
  filter(n > 1)

2.4 Cleaning the Data

Finally, we removed unnecessary columns automatically generated during import (e.g., those starting with ...1) and verified the structure and completeness of the cleaned dataset.

combined_data <- combined_data_beforeclean %>%
  select(-starts_with("...1"))

# Inspect structure and missing values
str(combined_data)
sum(is.na(combined_data))
colSums(is.na(combined_data))
combined_data %>% filter(if_any(everything(), is.na))

# Preview cleaned dataset
head(combined_data)

The resulting combined_data contains aligned observations for all variables of interest for the selected countries, covering the years 1800–2040. This verified and cleaned dataset serves as the basis for subsequent statistical analysis and visualization.

3.1 Time-based Line charts

We first created visualizations to explore basic time-series data across four countries. The graphs are organized to show key socio-economic and demographic trends from 1800 to 2040, in the following order:

• Gini Index – Illustrates shifts in income inequality.
  
• Life Expectancy at Birth – Shows improvements in health and longevity.
  
• Population Growth – Captures expansion patterns across racial and ethnic groups.
  
• Babies per Woman – Reflects demographic transition and declining birth rates.
• Household Income – Tracks the long-term trajectory of income growth across different groups.
  
• Average Daily Income per Person – Highlights changes in individual earning power over time.

These visualizations help us understand how income characteristics relate to fertility and living conditions in each country. When looking at these visualizations, around the 1800s to early 1900s, most factors indicate that these nations began under similar early-development conditions: extremely low income levels, high fertility rates, and short life expectancy. Household and daily income were generally minimal, with the majority of the population living near subsistence levels, often earning less than a few dollars per day. The birth rate remained high, averaging around six to seven babies per woman, which aligns with the characteristics of pre-demographic-transition societies where large families were necessary for labor and survival. Meanwhile, life expectancy hovered between 30 and 35 years, reflecting widespread poverty, limited healthcare access, and high infant mortality rates. The Gini index, representing income inequality, was relatively moderate across most countries (30-40), yet Tunisia stood out with an exceptionally high inequality level nearing 80, suggesting deep social or economic divides even at an early stage. Overall, this paints a baseline picture of four nations sharing low-income, high-fertility, and low-longevity beginnings, but with Tunisia already displaying a distinct imbalance in wealth distribution.

3.2 Normalized Diagrams of Each Country

In the next stage of our analysis, we applied Min-Max normalization to all input variables for each country, transforming them onto a unified 0–1 scale. This allowed us to visualize growth and decline patterns over time on a comparable basis. Specifically, variables such as average daily income, household income, Gini index, life expectancy, and babies per woman were normalized using the formula:

\[ x_{\text{norm}} = \frac{x - x_{\min}}{x_{\max} - x_{\min}} \]

where:
\(x\) = the original value of the variable
\(x_{\min}\) = minimum of the period means for that feature (over all periods)
\(x_{\max}\) = maximum of the period means for that feature (over all periods)
\(x_{\text{norm}}\) = the normalized value, scaled between 0 and 1

By standardizing variables in this way across a 50-year period, we removed the influence of differing units and magnitudes. This approach enables a clearer comparison of how each factor evolved relative to others within the same country. The resulting normalized values provide a balanced view of each country’s trajectory over time—highlighting periods of advancement, stagnation, or decline—and help identify key turning points in preparation for a deeper analysis of each country’s unique story.

According to four normalized diagrams, clear patterns of economic growth and social transformation of each country are obviously revealed. All countries experienced a steady increase in both average daily income and household income, along with a noticeable rise in life expectancy, particularly after 1950 - a period marked by modernization, improved infrastructure, and better access to healthcare. At the same time, fertility rates declined sharply, reflecting a demographic shift toward smaller families as living standards and education improved. When comparing countries, Egypt and Tunisia show strong upward momentum in both income and health outcomes, accompanied by a rapid fertility drop. Libya stands out with a dramatic post-1950 boom in income and life expectancy - driven largely by oil wealth - followed by a steep fall in birth rates. In contrast, Sudan’s growth was slower and more gradual, with high fertility persisting until the late 20th century before finally starting to decline.

3.3 Income Level Distribution Over Time

We chose to illustrate only the income level diagrams for each country across time because income serves as the core independent variable in our analysis. Our main research question focuses on how changes in financial inputs influence health outcomes. By isolating income trends first, we can clearly visualize the foundation of economic growth and identify when major transitions occurred (for example, post-1950 economic expansion). This step allows us to later connect these income shifts to health improvements and inequality patterns, making the relationship between wealth and well-being more explicit and connected.

How to Read the Graph
In the 1800s, almost everyone in the four countries belonged to Level 1 (dark blue) — representing extremely low income and limited access to basic needs.
Over time, especially after the 1900s, the light blue area (Level 2) expanded, meaning more people moved into slightly higher income levels.
By 1950–2040, the yellow area (Level 3) emerged and grew, indicating the rise of a middle-income population.
The orange and red areas (Levels 4 and 5) remained relatively small, suggesting that only a few people had reached the highest income levels.

Comparative Insights
In the early 1800s, most people in all four countries belonged to Level 1, representing extremely low income and limited access to basic needs. However, from the mid-20th century onward - particularly after the 1950s - there was a noticeable upward movement of population into higher income levels, reflecting rapid development, industrialization, and social modernization.

Among the four, Libya demonstrates the most dramatic income shift, with large proportions of its population moving into Levels 3–5 after the discovery of oil. Tunisia and Egypt show steadier, moderate transitions, gradually reducing their low-income population. Sudan, while improving, still retains a dominant share of lower-level income groups, signaling slower and more uneven growth.

3.4 Key Time Period: 1950s–2000s

The period from the 1950s to the early 2000s marks a crucial turning point for all four countries, as both income and health indicators rose sharply while fertility rates began to fall. This transformation reflects the broader wave of post-colonial nation-building and modernization across North Africa. Following independence — Egypt (1952), Sudan (1956), Tunisia (1956), and Libya (1951) — new governments launched national development programs aimed at economic self-sufficiency, healthcare expansion, and educational reform.

Key country-specific developments include:
- Egypt: Nasser’s socialist reforms and industrialization policies in the 1950s–60s helped raise living standards and expand public health coverage.
- Tunisia: Under Habib Bourguiba, heavy investments in women’s education and family planning led to one of the region’s most rapid fertility declines.
- Libya: After the discovery of oil in 1959, the economic boom funded hospitals, infrastructure, and welfare programs, sharply boosting both income and life expectancy.
- Sudan: Despite political instability, periods of agricultural growth and international aid during the 1970s–80s contributed to improvements in health and economic conditions.

Collectively, these changes pushed all four countries into the later stages of the demographic transition, defined by longer lifespans, smaller families, and the beginnings of modern middle-class societies.

4.1 Shared Income Level contributions

4.2 Average household income contributions

4.3 Gini Ranges contributions

4.4 Fertility rate and Life Expectancy

Babies Per Woman and Life Expectancy: Negative Relationship

The box plot presents a negative relationship between life expectancy at birth and fertility rate across countries. As life expectancy increases, the number of babies per woman declines notably. Countries within the two lowest life expectancy ranges (6-21 years and 21-37 years) display a stable fertility pattern, with women having around six to seven children on average and very little variation between nations. This suggests that fertility remains consistently high regardless of short lifespan. In contrast, as life expectancy rises beyond 37 years, fertility rates begin to drop sharply. Nations within the highest range (68-83 years) record the lowest fertility levels, averaging about two children per woman. The decreasing and stabilizing trend at higher life expectancy levels indicates that as countries develop and citizens live longer, family sizes tend to shrink to a lower rate.

5.1 Income Level and Babies Per Woman

5.2 Household Income and Life Expectancy

5.3 Household Income and Babies Per Woman

5.4 Gini And Babies Per Woman