R_datasci_ys_.Rmd

---
title: "R for Data Science Skill A."
author: "Yasemin Sucu"
date: "2023-11-16"
output:
  html_document:
    toc: true
    toc_depth: 3
    theme: readable 
    number_sections: yes
    toc_float:
      collapsed: yes
css: "styles.css"
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
```

Loading the libraries

```{r}
library(tidyverse)
library(plotly)
library(broom)
library(styler)
```

# Reading the CSV file into a tibble

```{r}
gapminder_data <- read.csv("/Users/yasemindilarasucu/Desktop/gapminder_data_ys/gapminder_clean.csv")
```

Filtering the Data by the year 1962

```{r}
filtered_data_1962 <- gapminder_data %>%
  filter(Year == 1962)
```

Creating a Scatter plot

```{r}
ggplot(filtered_data_1962, aes(x = " CO2 emissions (metric tons per capital) ", y = gdpPercap)) +
  geom_point() +
  labs(
    title = "Scatter Plot of CO2 Emissions vs. GDP per Capita (1962)",
    x = "CO2 Emissions (metric tons per capita)",
    y = "GDP per Capita"
  )
```

Checking for the missing values in the dataset

```{r}
summary(gapminder_data)
```

```{r}
summary(filtered_data_1962)
```

Imputing the NA values for the columns of CO2 Emissions (metric tons per capita and GDP per Capita

```{r}
filtered_data_1962_imputed <- filtered_data_1962 %>%
  mutate(
    "CO2 Emissions (metric tons per capita)" = ifelse(
      is.na("CO2 Emissions (metric tons per capita)"),
      mean("CO2 Emissions (metric tons per capita)", na.rm = TRUE),
      "CO2 Emissions (metric tons per capita)"
    ),
    gdpPercap = ifelse(
      is.na(gdpPercap),
      mean(gdpPercap, na.rm = TRUE),
      gdpPercap
    )
  )
```

#  Making a scatter plot with the modified dataset

```{r}
p <- ggplot(filtered_data_1962_imputed, aes(x = `CO2.emissions..metric.tons.per.capita.`, y = gdpPercap, color = continent)) +
  geom_point() +
  labs(
    title = "Scatter Plot of CO2 Emissions vs. GDP per Capita (1962)",
    x = "CO2 Emissions (metric tons per capita)",
    y = "GDP per Capita"
  )

#applying a theme 

p + theme_classic()

```

#  Calculating the correlation in between CO2 emissions (metric tons per capita) and gdpPercap, and p-value.

```{r}
correlation_result <- cor.test(filtered_data_1962_imputed$`CO2.emissions..metric.tons.per.capita.`,
  filtered_data_1962_imputed$gdpPercap,
  method = "pearson"
)

print(correlation_result)
```

#  In what year is the correlation between 'CO2 emissions (metric tons per capita)' and gdpPercap the strongest?

```{r}
# Finding the year with the strongest correlation

correlation_results <- lapply(unique(gapminder_data$Year), function(year) {
  subset_data <- gapminder_data %>% filter(Year == year)
  correlation_result <- cor.test(subset_data$`CO2.emissions..metric.tons.per.capita.`,
    subset_data$gdpPercap,
    method = "pearson"
  )
  return(data.frame(Year = year, correlation = correlation_result$estimate))
})

correlation_results_df <- do.call(rbind, correlation_results)
strongest_correlation_year <- correlation_results_df[which.max(correlation_results_df$correlation), "Year"]

# Filtering the data for the year of the strongest correlation

filtered_data_strongest_correlation <- gapminder_data %>% filter(Year == strongest_correlation_year)
```

#  Creating an interactive scatter plot with plotly for comparing 'CO2 emissions (metric tons per capita)' and gdpPercap.

```{r}
# Creating a scatter-plot with ggplot2 with the filtered data from the previous step

gg_plot <- ggplot(filtered_data_strongest_correlation, aes(x = `CO2.emissions..metric.tons.per.capita.`, y = gdpPercap, size = pop, color = continent)) +
  geom_point() +
  labs(
    title = "Interactive Scatter Plot with Plotly",
    x = "CO2 Emissions (metric tons per capita)",
    y = "GDP per Capita"
  )

# Converting ggplot2 plot to plotly

plotly_plot <- ggplotly(gg_plot)

plotly_plot
```

# Questions

1.  What's the relationship in between the continent and ' Energy use (kg of oil equivalent per capita)' ?

```{r}
# Checking the significancy with ANOVA test

anova_result <- aov(`Energy.use..kg.of.oil.equivalent.per.capita.` ~ continent, data = gapminder_data)

summary(anova_result)
```

Commenting on the results that the analysis shows that there are significant differences in Energy use among the different continents.

Visualization:

```{r}
# Interactive Boxplot

plot_ly(gapminder_data, x = ~continent, y = ~`Energy.use..kg.of.oil.equivalent.per.capita.`, type = "box") %>%
  layout(
    title = "Energy Use Across Continents",
    xaxis = list(title = "Continent"),
    yaxis = list(title = "Energy Use (kg of oil equivalent per capita)")
  )
```

2.  Is there a significant difference between Europe and Asia with respect to 'Imports of goods and services (% of GDP)' in the years after 1990?

```{r}
# Checking unique levels of the 'continent' variable
unique(gapminder_data$continent)
```

```{r}
# Filter data for non-empty continents and years after 1990
filtered_data_after_1990_cleaned <- gapminder_data %>%
  filter(Year > 1990, continent %in% c("Asia", "Europe", "Africa", "Americas", "Oceania"))

# Explicitly set factor levels for 'continent'
filtered_data_after_1990_cleaned$continent <- factor(filtered_data_after_1990_cleaned$continent, levels = c("Asia", "Europe"))

# Check unique levels of the 'continent' variable
unique(filtered_data_after_1990_cleaned$continent)
```

```{r}
# Using the t-test for to compare the means of "Import of goods and services (% of GDP)" between Europe and Asia after 1990

t_test_result <- t.test(`Imports.of.goods.and.services....of.GDP.` ~ continent, data = filtered_data_after_1990_cleaned)

t_test_result
```

#According the t-test results that there is no significant difference in the means of "Import of goods and services (% of GDP)" between Europe and Asia after 1990, the p-value is 0.1776, which is greater than the typical significance level of 0.05. Therefore, there is no enough evidence to reject the null hypothesis that true difference in means between Asia and Europe is equal to 0.

Visualization

```{r}
# Interactive Bar Plot
plot_ly(filtered_data_after_1990_cleaned, x = ~continent, y = ~`Imports.of.goods.and.services....of.GDP.`, type = "bar") %>%
  layout(
    title = "Imports of Goods and Services (% of GDP) - Europe vs. Asia (After 1990)",
    xaxis = list(title = "Continent"),
    yaxis = list(title = "Imports (% of GDP)")
  )
```

3.  What is the country (or countries) that has the highest 'Population density (people per sq. km of land area)' across all years? (i.e., which country has the highest average ranking in this category across each time point in the dataset?)

```{r}
# Since there is no need for a statistical test for this question, instead let's calculate the Population Density for each country across all years and identify the one with the highest average.

# Calculate average 'Population Density' for each country
average_population_density <- gapminder_data %>%
  group_by(Country.Name) %>%
  summarize(avg_population_density = mean(`Population.density..people.per.sq..km.of.land.area.`, na.rm = TRUE)) %>%
  arrange(desc(avg_population_density))

# Display the country with the highest average population density
head(average_population_density, 1)
```

Visualization

```{r}
# Interactive Bar Plot
plot_ly(average_population_density, x = ~Country.Name, y = ~avg_population_density, type = "bar") %>%
  layout(
    title = "Average Population Density Across All Years",
    xaxis = list(title = "Country"),
    yaxis = list(title = "Average Population Density")
  )
```

4.  What country (or countries) has shown the greatest increase in 'Life expectancy at birth, total (years)' between 1962 and 2007?

```{r}
# Since there is no need for statistical test for this one, let's calculate the difference in "Life expectancy in between 1962 and 2007 for each country and identifying the one with he greatest increase.

# Calculate the difference in life expectancy between 1962 and 2007 for each country
life_expectancy_difference <- gapminder_data %>%
  filter(Year %in% c(1962, 2007)) %>%
  group_by(Country.Name) %>%
  summarize(difference = diff(`Life.expectancy.at.birth..total..years.`)) %>%
  arrange(desc(difference))

# Display the country with the greatest increase in life expectancy
head(life_expectancy_difference, 1)
```

Visualization

```{r}
# Create an interactive scatter plot
plot_ly(
  data = life_expectancy_difference, x = ~difference, y = ~Country.Name, type = "scatter", mode = "markers",
  marker = list(color = ~difference, colorscale = "Viridis"),
  text = ~ paste("Country: ", Country.Name, "<br>Life Expectancy Difference: ", round(difference, 2))
) %>%
  layout(
    title = "Difference in Life Expectancy (2007 - 1962) for Each Country",
    xaxis = list(title = "Life Expectancy Difference"),
    yaxis = list(title = "Country"),
    hovermode = "closest"
  )
```