Apply it to your data 6

Import Data

outer_space_objects <- readr::read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2024/2024-04-23/outer_space_objects.csv')

## Rows: 1175 Columns: 4
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (2): Entity, Code
## dbl (2): Year, num_objects
## 
## 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.

Tidy data

set.seed(2) # for reproducible outcome
data_top10_launchers <- outer_space_objects %>%
    
    # Select three columns
    select(Entity, Year, num_objects ) %>%
    group_by(Entity) %>%
    filter(num_objects == max(num_objects)) %>%
    ungroup() %>%
    arrange(desc(num_objects)) %>%
    slice_head(n = 10)
    
print(data_top10_launchers)

## # A tibble: 10 × 3
##    Entity          Year num_objects
##    <chr>          <dbl>       <dbl>
##  1 World           2023        2664
##  2 United States   2023        2166
##  3 United Kingdom  2021         289
##  4 China           2022         182
##  5 Russia          1981         124
##  6 Belgium         2017          28
##  7 Japan           2014          24
##  8 Japan           2021          24
##  9 France          2011          19
## 10 Spain           2022          19

Introduction

Questions

Variation

Visualizing distributions

Bar Plot

data_top10_launchers %>%
  ggplot(aes(x = reorder(Entity, -num_objects), y = num_objects, fill = factor(Year))) +
  geom_bar(stat = "identity", position = "dodge") +
  labs(title = "Top Space Object Launches Globally and by Country", x = "Entity", y = "Number of Objects Launched") +
  scale_y_continuous(breaks = seq(0, max(data_top10_launchers$num_objects), by = 250)) +
  theme(axis.text.x = element_text(angle = 45, hjust = 1, vjust = 1), 
        axis.ticks.x = element_blank())

Unusual values

ScatterPlot that shows that The United States makes almost all of the launches that contribute to the total World count, which is a significant outlier in the dataset

data_top10_launchers %>%
  ggplot(aes(x = Year, y = num_objects, color = Entity)) +
  geom_point() +
  labs(title = "Space Object Launches Over the Years", x = "Year", y = "Number of Objects Launched") +
  scale_y_continuous(breaks = seq(0, max(data_top10_launchers$num_objects), by = 250)) +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

Missing Values

I have set a limiter on the previous tables y axis count to remove outliers with launches under 5 for each year and over 2000 for each year, this removed our outlier values for the United States

library(tidyverse)

data_top10_launchers %>% 
  mutate(num_objects = ifelse(num_objects < 5 | num_objects > 2000, NA, num_objects)) %>% 
  ggplot(aes(x = Year, y = num_objects, color = Entity)) + 
  geom_point() +
  scale_y_continuous(breaks = seq(0, max(data_top10_launchers$num_objects, na.rm = TRUE), by = 25)) +
  theme(axis.text.x = element_text(angle = 45, hjust = 1)) +
  labs(title = "Space Object Launches with Outliers Removed", x = "Year", y = "Number of Objects Launched")

## Warning: Removed 2 rows containing missing values or values outside the scale range
## (`geom_point()`).

Covariation

A categorical and continuous variable

library(tidyverse)

data_top10_launchers %>%
  ggplot(aes(x = reorder(Entity, -num_objects), y = num_objects, fill = Entity)) +
  geom_col() +
  labs(title = "Number of Space Objects Launched by Country",
       x = "Entity",
       y = "Number of Objects Launched") +
  scale_y_continuous(breaks = seq(0, max(data_top10_launchers$num_objects), by = 250)) +
  theme_minimal() +
  theme(axis.text.x = element_text(angle = 45, hjust = 1), legend.position = "none")

Two categorical variables

Isn’t applicable for this dataset, my tidied set has one categorical, the full set had two but it wouldn’t give any insight because they were just country names in the ‘entity’ column and their abbreviations in a ‘Code’ column.

Two continous variables

library(hexbin)
library(tidyverse)
library(hexbin)

data_top10_launchers %>% 
  ggplot(aes(x = Year, y = num_objects)) + 
  geom_hex() +
  labs(title = "Hexbin Plot of Year vs. Number of Objects Launched",
       x = "Year",
       y = "Number of Objects Launched") +
  theme_minimal()

Patterns and models

This is a scatterplot utilizing a color coding scheme for each entity and what the red line indicated is the modelled increase in object launches every year, as can be seen theres overall a positive nonlinear correlation in the model shown.

# Building the polynomial model
mod_poly <- lm(num_objects ~ poly(Year, 2), data = data_top10_launchers)

# Adding residuals to the data
data_top10_with_resid <- data_top10_launchers %>%
  modelr::add_residuals(mod_poly)


# Plotting the fitted polynomial model with color-coded dots
data_top10_launchers %>%
  ggplot(aes(x = Year, y = num_objects, color = Entity)) +
  geom_point() +
  geom_smooth(method = "lm", formula = y ~ poly(x, 2), se = FALSE, color = "red") +
  labs(title = "Number of Space Objects Launched Over the Years",
       x = "Year",
       y = "Number of Objects Launched") +
  theme_minimal() +
  scale_y_continuous(breaks = seq(0, max(data_top10_launchers$num_objects), by = 250)) +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))