| Country | LifeExpectancy |
|---|---|
| Monaco | 77.25278 |
| Andorra | 77.04861 |
| San Marino | 76.88056 |
| Guernsey | 76.71944 |
| Israel | 75.49722 |
Data Visualization with ggplot2
From Points and Bars to Boxplots and Maps in ggplot2
Bogdan G. Popescu 
John Cabot University
Why Visuals?
Numbers vs Visuals
Would you rather read this?
…or see this?
The ggplot2 library in R
R can help us achieve great visualizations with the help of ggplot2
The ggplot2 library in R
R can help us achieve great visualizations with the help of ggplot2
The ggplot2 library in R
R can help us achieve great visualizations with the help of ggplot2
The Essentials
Grammar of Graphics
At the most basic level we have:
- data
- geometries
- aesthetics
Optional layers include:
- facets
- coordinates
- annotations
- themes, etc.
- We can add them together in
ggplot()with+
Data Cleaning
Preparing the Data
Before we use ggplot, our data has to be tidy. This means:
- Each variable is a column
- Each observation has its own row
- Each value has its own cell
Simple Geoms
Geometries
Simple Geoms
Geometries
Geoms are the “shapes” your data takes on a graph.
Example:
- points (geom_point) make a scatterplot
- bars (geom_bar) make a bar chart
- lines (geom_line) make a line graph.
Simple Geoms
Examples
geom_point
geom_col
geom_bar
geom_text
geom_lineSimple Geoms
Reference
geom_point
geom_col
geom_bar
geom_text
geom_line| geom | Use for |
|---|---|
geom_point() |
Relationships between two variables; at least 10 obs. |
geom_col() |
Totals/percent per category; ordered bars. Uses pre-computed values for bar height. You must supply both x and y |
geom_bar() |
Counts the observations in each category. You must supply x, not y |
geom_text() |
Direct labels for small N; annotate outliers |
geom_line() |
Relationships between two variables; at least 10 obs. |
Simple Geoms
geom_point
Imagine we have data about 9 countries that records their level of democracy from -10 to +10 (x-axis) and their GDP per capita in $1,000s (y-axis).
Simple Geoms
geom_point
Interpretation: Each dot shows one country’s democracy score (left to right) and its GDP per person (up and down). The dots suggest that more democratic countries usually have higher GDP per person.
Simple Geoms
geom_col
Suppose we collected a small survey and just recorded the education level of each respondent. We want to visualize how many respondents fall into each category.
Simple Geoms
geom_bar
Suppose we collected a small survey and just recorded the education level of each respondent. We want to visualize how many respondents fall into each category.
What is the difference?
geom_col vs. geom_bar
This is the difference:
Simple Geoms
geom_text
Suppose we collected a small survey and just recorded the education level of each respondent. We want to visualize how many respondents fall into each category.
Simple Geoms
geom_text
We can also use geom_text with the first example:
Imagine we have data about 9 countries that records their level of democracy from -10 to +10 (x-axis) and their GDP per capita in $1,000s (y-axis).
Simple Geoms
geom_line
Suppose we have data on average voter turnout (%) in national elections over several years. We want to see the trend in participation.
Complex Geoms
Examples
geom_boxplot
geom_histogram
geom_density
geom_violin
geom_smooth
geom_errorbar
geom_sfComplex Geoms
Reference
geom_boxplot
geom_histogram
geom_density
geom_violin
geom_smooth
geom_errorbar
geom_sf| geom | Use for |
|---|---|
geom_boxplot(). |
Compare distributions across groups |
geom_histogram() |
Distribution of a single continuous variable (frequency bins) |
geom_density() |
Smoothed distribution of a single continuous variable |
geom_violin() |
Visualizing distributions across categories (shape of data, not just box) |
geom_smooth() |
Adding fitted trend/smoothed lines (LOESS, GAM, linear, etc.) |
geom_errorbar() |
Showing uncertainty or variability (e.g., CI ranges around a mean) |
geom_sf() |
Plotting spatial (map) data stored as sf objects |
Complex Geoms
geom_boxplot
Suppose we surveyed people about their trust in government on a 1–10 scale (1 = no trust, 10 = complete trust). We want to see the typical values and how spread out the answers are.
Complex Geoms
geom_boxplot
Suppose we surveyed people about their trust in government on a 1–10 scale (1 = no trust, 10 = complete trust). We want to see the typical values and how spread out the answers are.
Complex Geoms
geom_boxplot
Suppose we surveyed people about their trust in government on a 1–10 scale (1 = no trust, 10 = complete trust). We want to compare typical values and how spread out the answers are for men and women.
Complex Geoms
geom_boxplot
Suppose we surveyed people about their trust in government on a 1–10 scale (1 = no trust, 10 = complete trust). We want to compare typical values and how spread out the answers are for men and women.
Complex Geoms
geom_histogram: What is a Histogram?
Histograms allow us to better understand how frequently or infrequently certain values occur in our dataset.
Imagine a set of values that are spaced out along a number line.
Complex Geoms
geom_histogram: What is a Histogram?
To construct a histogram, a section of the number line is divided into equal chunks, called bins.
Next, count how many data points sit inside each bin, and draw bars, one for each bin
The heights of the bars correspond to the number of data points.
Complex Geoms
geom_histogram: What is a Histogram?
Label the data (in the example below each data point is an SAT score)
Draw in a y-axis which counts the number of data points in each bin
Finally label your bins.
Complex Geoms
geom_histogram: Example
Suppose we surveyed people about their trust in government on a 1–10 scale (1 = no trust, 10 = complete trust). We want to compare typical values and how spread out the answers are.
Complex Geoms
geom_histogram: Example
Suppose we surveyed people about their trust in government on a 1–10 scale (1 = no trust, 10 = complete trust). We want to compare typical values and how spread out the answers are.
There is a direct connection between histograms and boxplots:
Complex Geoms
geom_histogram: Example
Suppose we surveyed people about their trust in government on a 1–10 scale (1 = no trust, 10 = complete trust). We want to compare typical values and how spread out the answers are.
There is a direct connection between histograms and boxplots:
Complex Geoms
geom_histogram: Example
One specific aspect pertaining to histograms is the number of bins:
- too wide bins hide structure;
- too narrow bins are noisy..
Complex Geoms
geom_density: What is a Density Curve?
A histogram counts how many students fall in each trust “bucket.”
Complex Geoms
geom_density: What is a Density Curve?
A histogram counts how many students fall in each trust “bucket.”
A density curve smooths those buckets into a single curve so we see the overall shape.
Complex Geoms
geom_density: What is a Density Curve?
A histogram counts how many students fall in each trust “bucket.”
A density curve smooths those buckets into a single curve so we see the overall shape.
The area under the entire curve = 1 (i.e., 100% of people).
Complex Geoms
geom_density: What is a Density Curve?
A histogram counts how many students fall in each trust “bucket.”
A density curve smooths those buckets into a single curve so we see the overall shape.
The area under the entire curve = 1 (i.e., 100% of people).
The probability that a randomly chosen person’s trust score is between certain values (e.g. 5 and 7)
Complex Geoms
geom_density: What is a Density Curve?
A histogram counts how many students fall in each trust “bucket.”
A density curve smooths those buckets into a single curve so we see the overall shape.
The area under the entire curve = 1 (i.e., 100% of people).
The probability that a randomly chosen person’s trust score is above 5.
Complex Geoms
geom_density: What is a Density Curve?
A histogram counts how many students fall in each trust “bucket.”
A density curve smooths those buckets into a single curve so we see the overall shape.
The area under the entire curve = 1 (i.e., 100% of people).
The probability that a randomly chosen person’s trust score is below 4.
Complex Geoms
geom_density: What is a Density Curve?
Heights tell you nothing by themselves—areas tell you the share.
When I widen the range, the area grows—so the probability grows.
This curve is smoothed from the histogram; bandwidth controls how smooth.
Complex Geoms
geom_density: What is a Density Curve?
Heights tell you nothing by themselves—areas tell you the share.
When I widen the range, the area grows—so the probability grows.
This curve is smoothed from the histogram; bandwidth controls how smooth.
Here too there is a connection between density, histograms, and boxplots.
Complex Geoms
geom_density: Example
Suppose we surveyed people about their trust in government on a 1–10 scale (1 = no trust, 10 = complete trust). We want to compare typical values and how spread out the answers are.
Complex Geoms
geom_density: Example
Suppose we surveyed people about their trust in government on a 1–10 scale (1 = no trust, 10 = complete trust). We want to compare typical values and how spread out the answers are.
It is more informative to include both geom_density and geom_histogram.
Show the code
# Match density to histogram counts
N <- nrow(eg5)
binwidth <- diff(range(eg5$trust)) / 10 # because bins = 10
# Draw histogram
ggplot(eg5, aes(x = trust)) +
geom_histogram(bins = 10, color = "white") +
geom_density(aes(y = after_stat(density) * N * binwidth), linewidth = 1) +
labs(
x = "trust",
y = "count")
Complex Geoms
geom_density: Example
Most people in the survey gave middle-of-the-road trust scores—think around 5 to 7.
As you move toward the extremes (very low 1–3 or very high 8–10), the curve drops, showing fewer people chose those scores.
Show the code
# Match density to histogram counts
N <- nrow(eg5)
binwidth <- diff(range(eg5$trust)) / 10 # because bins = 10
# Draw histogram
ggplot(eg5, aes(x = trust)) +
geom_histogram(bins = 10, color = "white") +
geom_density(aes(y = after_stat(density) * N * binwidth), linewidth = 1) +
labs(
x = "trust",
y = "count")Complex Geoms
geom_violin: Example
Violin plots are symmetric representations of geom_density
Complex Geoms
geom_violin: Example
Violin plots are symmetric representations of geom_density
This was our density
Complex Geoms
geom_violin: Example
Violin plots are symmetric representations of geom_density
This was our density shaded in white
Complex Geoms
geom_violin: Example
Violin plots are symmetric representations of geom_density
This is our density flipped
Complex Geoms
geom_violin: Example
Violin plots are symmetric representations of geom_density
This is what happens if we create a mirror density
Complex Geoms
geom_violin: Example
Complex Geoms
geom_violin: Example
Suppose we surveyed people about their trust in government on a 1–10 scale (1 = no trust, 10 = complete trust). We want to compare typical values and how spread out the answers are for men and women.
Complex Geoms
geom_violin: Example
Interpretation: The data suggests that men’s responses are more spread out (from low to high trust), while women’s are more concentrated in the middle-to-high range.
Complex Geoms
geom_smooth
This is what we had previously:
Complex Geoms
geom_smooth
This is what happens when we try to fit a line:
Complex Geoms
geom_smooth
Interpretation: As one moves right (more democratic), the dots usually sit higher—meaning those countries tend to be richer. The black line going up summarizes that big-picture pattern, even though a few dots don’t fit.
Complex Geoms
geom_errorbar
Error bars are like “antennae” on top of a mean.
They don’t show the full spread like a boxplot, but they give us a quick picture of how precise the average is.
- Short bars → data points are close together.
- Long bars → data points are spread out.
This is useful when we only want to compare averages.
Complex Geoms
geom_errorbar
A natural question is how do error bars compare to boxplots.
Boxplot
- Shows the distribution of the data: Median, quartiles, and possible outliers.
- Good for: seeing the shape and spread
Error Bar
- Summarizes just the mean + uncertainty (often mean ± standard error or confidence interval).
- Doesn’t show the full shape of the data.
Complex Geoms
geom_errorbar
Suppose we surveyed people about their trust in government on a 1–10 scale (1 = no trust, 10 = complete trust). We want to compare typical values and how spread out the answers are for men and women.
# Set a random seed so results are reproducible.
# (Without this, R will generate different random numbers each time.)
set.seed(123)
# -------------------------------------------------------------
# Step 1: Create a toy dataset: "trust in government" by gender
# --------------------------------------------------------------
eg5 <- data.frame(
# Repeat the labels "Men" and "Women" 20 times each
gender = rep(c("Men", "Women"), each = 20),
# Generate 20 trust values for Men ~ Normal(mean=5, sd=2)
# Generate 20 trust values for Women ~ Normal(mean=8, sd=1)
trust = c(
rnorm(20, mean = 5, sd = 2),
rnorm(20, mean = 8, sd = 1)
)
)
#Print the first 3 entries
head(eg5, n=3)| gender | trust |
|---|---|
| Men | 3.879049 |
| Men | 4.539645 |
| Men | 8.117417 |
Complex Geoms
geom_errorbar
Suppose we surveyed people about their trust in government on a 1–10 scale (1 = no trust, 10 = complete trust). We want to compare typical values and how spread out the answers are for men and women.
# ----------------------------------------------------------------------
# Step 2: Calculate mean trust and error bars (95% confidence intervals)
# ----------------------------------------------------------------------
df_err <- eg5 %>%
group_by(gender) %>% # Group data by gender
summarise(
mean = mean(trust), # Average trust per gender
se = sd(trust) / sqrt(n()), # Standard error = sd / sqrt(sample size)
) %>%
mutate(
# 95% confidence interval = mean ± 1.96 * standard error
ymin = mean - 1.96 * se, # Lower bound
ymax = mean + 1.96 * se # Upper bound
)
head(df_err, n=3)| gender | mean | se | ymin | ymax |
|---|---|---|---|---|
| Men | 5.283248 | 0.4349892 | 4.430669 | 6.135826 |
| Women | 7.948743 | 0.1855799 | 7.585006 | 8.312480 |
Complex Geoms
geom_errorbar
Suppose we surveyed people about their trust in government on a 1–10 scale (1 = no trust, 10 = complete trust). We want to compare typical values and how spread out the answers are for men and women.
# --------------------------------------------------------
# Step 3: Plotting the means with error bars using ggplot2
# --------------------------------------------------------
ggplot(df_err, aes(x = gender, y = mean)) +
geom_point(size = 4) + # Plot mean values as large points
geom_errorbar(
aes(ymin = ymin, ymax = ymax), # Add vertical error bars
width = 0.15 # Small horizontal "cap" on error bars
) +
labs(
x = NULL, # Remove x-axis label
y = "trust", # Label y-axis
title = "Error Bars" # Add plot title
) +
coord_cartesian(ylim = c(0, 10)) # Set y-axis limits from 0 to 10
Complex Geoms
geom_errorbar
Suppose we surveyed people about their trust in government on a 1–10 scale (1 = no trust, 10 = complete trust). We want to compare typical values and how spread out the answers are for men and women.
Show the code
set.seed(123)
#Step1: Toy dataset: trust in government by gender
eg5 <- data.frame(
gender = rep(c("Men", "Women"), each = 20),
trust = c(
rnorm(20, mean = 5, sd = 2),
rnorm(20, mean = 8, sd = 1)))
#Step2: Calculating Error bars
df_err <- eg5 %>%
group_by(gender) %>%
summarise(
mean = mean(trust),
se = sd(trust) / sqrt(n()),
) %>%
mutate(
ymin = mean - 1.96 * se, # 95% CI lower bound
ymax = mean + 1.96 * se # 95% CI upper bound
)
#Step3: Plotting
ggplot(df_err, aes(x = gender, y = mean)) +
geom_point(size = 4) +
geom_errorbar(aes(ymin = ymin, ymax = ymax), width = 0.15) +
labs(
x = NULL,
y = "trust",
title = "Error Bars"
) +
coord_cartesian(ylim = c(0, 10)) # same y-axisComplex Geoms
geom_errorbar
Interpretation: Women in this sample show higher average trust in government than men. The bars show our uncertainty, and since they don’t overlap much, it means the difference is likely real and not just due to chance.
Show the code
set.seed(123)
#Step1: Toy dataset: trust in government by gender
eg5 <- data.frame(
gender = rep(c("Men", "Women"), each = 20),
trust = c(
rnorm(20, mean = 5, sd = 2),
rnorm(20, mean = 8, sd = 1)))
#Step2: Calculating Error bars
df_err <- eg5 %>%
group_by(gender) %>%
summarise(
mean = mean(trust),
se = sd(trust) / sqrt(n()),
) %>%
mutate(
ymin = mean - 1.96 * se, # 95% CI lower bound
ymax = mean + 1.96 * se # 95% CI upper bound
)
#Step3: Plotting
ggplot(df_err, aes(x = gender, y = mean)) +
geom_point(size = 4) +
geom_errorbar(aes(ymin = ymin, ymax = ymax), width = 0.15) +
labs(
x = NULL,
y = "trust",
title = "Error Bars"
) +
coord_cartesian(ylim = c(0, 10)) # same y-axisComplex Geoms
geom_sf
geom_sf draws maps from sf objects (“simple features” = data + geometry).
It works like other geoms: points and lines
Complex Geoms
geom_sf
geom_sf draws maps from sf objects (“simple features” = data + geometry).
With geom_sf, you can map points (cities, places, settlements)
Complex Geoms
geom_sf
geom_sf draws maps from sf objects (“simple features” = data + geometry).
With geom_sf, you can map lines (rivers, roads, straight lines)
Complex Geoms
geom_sf
geom_sf draws maps from sf objects (“simple features” = data + geometry).
With geom_sf, you can map polygons (countries, parks, seas, etc.)
Complex Geoms
geom_sf
geom_sf draws maps from sf objects (“simple features” = data + geometry).
It works like other geoms: points and lines
Typical workflow:
- Get a shape (countries, regions, cities, rivers, roads, etc.)
- Plot them with
geom_sf
Complex Geoms
geom_sf Example
# Install once (uncomment if needed):
# install.packages(c("sf", "ggplot2", "rnaturalearth", "rnaturalearthdata"))
library(ggplot2)
library(sf)
library(rnaturalearth)
library(rnaturalearthdata)
# Step1: Get a country polygons dataset as an sf object
world <- rnaturalearth::ne_countries(scale = "medium",
returnclass = "sf")
head(world, n=3)
| featurecla | scalerank | labelrank | sovereignt | sov_a3 | adm0_dif | level | type | tlc | admin | adm0_a3 | geou_dif | geounit | gu_a3 | su_dif | subunit | su_a3 | brk_diff | name | name_long | brk_a3 | brk_name | brk_group | abbrev | postal | formal_en | formal_fr | name_ciawf | note_adm0 | note_brk | name_sort | name_alt | mapcolor7 | mapcolor8 | mapcolor9 | mapcolor13 | pop_est | pop_rank | pop_year | gdp_md | gdp_year | economy | income_grp | fips_10 | iso_a2 | iso_a2_eh | iso_a3 | iso_a3_eh | iso_n3 | iso_n3_eh | un_a3 | wb_a2 | wb_a3 | woe_id | woe_id_eh | woe_note | adm0_iso | adm0_diff | adm0_tlc | adm0_a3_us | adm0_a3_fr | adm0_a3_ru | adm0_a3_es | adm0_a3_cn | adm0_a3_tw | adm0_a3_in | adm0_a3_np | adm0_a3_pk | adm0_a3_de | adm0_a3_gb | adm0_a3_br | adm0_a3_il | adm0_a3_ps | adm0_a3_sa | adm0_a3_eg | adm0_a3_ma | adm0_a3_pt | adm0_a3_ar | adm0_a3_jp | adm0_a3_ko | adm0_a3_vn | adm0_a3_tr | adm0_a3_id | adm0_a3_pl | adm0_a3_gr | adm0_a3_it | adm0_a3_nl | adm0_a3_se | adm0_a3_bd | adm0_a3_ua | adm0_a3_un | adm0_a3_wb | continent | region_un | subregion | region_wb | name_len | long_len | abbrev_len | tiny | homepart | min_zoom | min_label | max_label | label_x | label_y | ne_id | wikidataid | name_ar | name_bn | name_de | name_en | name_es | name_fa | name_fr | name_el | name_he | name_hi | name_hu | name_id | name_it | name_ja | name_ko | name_nl | name_pl | name_pt | name_ru | name_sv | name_tr | name_uk | name_ur | name_vi | name_zh | name_zht | fclass_iso | tlc_diff | fclass_tlc | fclass_us | fclass_fr | fclass_ru | fclass_es | fclass_cn | fclass_tw | fclass_in | fclass_np | fclass_pk | fclass_de | fclass_gb | fclass_br | fclass_il | fclass_ps | fclass_sa | fclass_eg | fclass_ma | fclass_pt | fclass_ar | fclass_jp | fclass_ko | fclass_vn | fclass_tr | fclass_id | fclass_pl | fclass_gr | fclass_it | fclass_nl | fclass_se | fclass_bd | fclass_ua | geometry |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Admin-0 country | 1 | 3 | Zimbabwe | ZWE | 0 | 2 | Sovereign country | 1 | Zimbabwe | ZWE | 0 | Zimbabwe | ZWE | 0 | Zimbabwe | ZWE | 0 | Zimbabwe | Zimbabwe | ZWE | Zimbabwe | NA | Zimb. | ZW | Republic of Zimbabwe | NA | Zimbabwe | NA | NA | Zimbabwe | NA | 1 | 5 | 3 | 9 | 14645468 | 14 | 2019 | 21440 | 2019 | 5. Emerging region: G20 | 5. Low income | ZI | ZW | ZW | ZWE | ZWE | 716 | 716 | 716 | ZW | ZWE | 23425004 | 23425004 | Exact WOE match as country | ZWE | NA | ZWE | ZWE | ZWE | ZWE | ZWE | ZWE | ZWE | ZWE | ZWE | ZWE | ZWE | ZWE | ZWE | ZWE | ZWE | ZWE | ZWE | ZWE | ZWE | ZWE | ZWE | ZWE | ZWE | ZWE | ZWE | ZWE | ZWE | ZWE | ZWE | ZWE | ZWE | ZWE | -99 | -99 | Africa | Africa | Eastern Africa | Sub-Saharan Africa | 8 | 8 | 5 | -99 | 1 | 0 | 2.5 | 8 | 29.92544 | -18.91164 | 1159321441 | Q954 | زيمبابوي | জিম্বাবুয়ে | Simbabwe | Zimbabwe | Zimbabue | زیمبابوه | Zimbabwe | Ζιμπάμπουε | זימבבואה | ज़िम्बाब्वे | Zimbabwe | Zimbabwe | Zimbabwe | ジンバブエ | 짐바브웨 | Zimbabwe | Zimbabwe | Zimbábue | Зимбабве | Zimbabwe | Zimbabve | Зімбабве | زمبابوے | Zimbabwe | 津巴布韦 | 辛巴威 | Admin-0 country | NA | Admin-0 country | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | MULTIPOLYGON (((31.28789 -2... |
| Admin-0 country | 1 | 3 | Zambia | ZMB | 0 | 2 | Sovereign country | 1 | Zambia | ZMB | 0 | Zambia | ZMB | 0 | Zambia | ZMB | 0 | Zambia | Zambia | ZMB | Zambia | NA | Zambia | ZM | Republic of Zambia | NA | Zambia | NA | NA | Zambia | NA | 5 | 8 | 5 | 13 | 17861030 | 14 | 2019 | 23309 | 2019 | 7. Least developed region | 4. Lower middle income | ZA | ZM | ZM | ZMB | ZMB | 894 | 894 | 894 | ZM | ZMB | 23425003 | 23425003 | Exact WOE match as country | ZMB | NA | ZMB | ZMB | ZMB | ZMB | ZMB | ZMB | ZMB | ZMB | ZMB | ZMB | ZMB | ZMB | ZMB | ZMB | ZMB | ZMB | ZMB | ZMB | ZMB | ZMB | ZMB | ZMB | ZMB | ZMB | ZMB | ZMB | ZMB | ZMB | ZMB | ZMB | ZMB | ZMB | -99 | -99 | Africa | Africa | Eastern Africa | Sub-Saharan Africa | 6 | 6 | 6 | -99 | 1 | 0 | 3.0 | 8 | 26.39530 | -14.66080 | 1159321439 | Q953 | زامبيا | জাম্বিয়া | Sambia | Zambia | Zambia | زامبیا | Zambie | Ζάμπια | זמביה | ज़ाम्बिया | Zambia | Zambia | Zambia | ザンビア | 잠비아 | Zambia | Zambia | Zâmbia | Замбия | Zambia | Zambiya | Замбія | زیمبیا | Zambia | 赞比亚 | 尚比亞 | Admin-0 country | NA | Admin-0 country | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | MULTIPOLYGON (((30.39609 -1... |
| Admin-0 country | 1 | 3 | Yemen | YEM | 0 | 2 | Sovereign country | 1 | Yemen | YEM | 0 | Yemen | YEM | 0 | Yemen | YEM | 0 | Yemen | Yemen | YEM | Yemen | NA | Yem. | YE | Republic of Yemen | NA | Yemen | NA | NA | Yemen, Rep. | NA | 5 | 3 | 3 | 11 | 29161922 | 15 | 2019 | 22581 | 2019 | 7. Least developed region | 4. Lower middle income | YM | YE | YE | YEM | YEM | 887 | 887 | 887 | RY | YEM | 23425002 | 23425002 | Exact WOE match as country | YEM | NA | YEM | YEM | YEM | YEM | YEM | YEM | YEM | YEM | YEM | YEM | YEM | YEM | YEM | YEM | YEM | YEM | YEM | YEM | YEM | YEM | YEM | YEM | YEM | YEM | YEM | YEM | YEM | YEM | YEM | YEM | YEM | YEM | -99 | -99 | Asia | Asia | Western Asia | Middle East & North Africa | 5 | 5 | 4 | -99 | 1 | 0 | 3.0 | 8 | 45.87438 | 15.32823 | 1159321425 | Q805 | اليمن | ইয়েমেন | Jemen | Yemen | Yemen | یمن | Yémen | Υεμένη | תימן | यमन | Jemen | Yaman | Yemen | イエメン | 예멘 | Jemen | Jemen | Iémen | Йемен | Jemen | Yemen | Ємен | یمن | Yemen | 也门 | 葉門 | Admin-0 country | NA | Admin-0 country | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | MULTIPOLYGON (((53.08564 16... |
Complex Geoms
geom_sf Example
# Install once (uncomment if needed):
# install.packages(c("sf", "ggplot2", "rnaturalearth", "rnaturalearthdata"))
library(ggplot2)
library(sf)
library(rnaturalearth)
library(rnaturalearthdata)
# Step1: Get a country polygons dataset as an sf object
world <- rnaturalearth::ne_countries(scale = "medium",
returnclass = "sf")
head(world, n=3)
So, the world dataframe contains 169 variables (e.g. featurecla, scalerank, labelrank, etc.) and 242 observations
Complex Geoms
geom_sf Example
Complex Geoms
geom_sf Example
Use coord_sf(xlim=..., ylim=...) to zoom
You can layer multiple sf objects:
- Countries (polygons)
- Cities (points)
Complex Geoms
geom_sf Example
Complex Geoms
geom_sf Example
Complex Geoms
geom_sf Example
europe_bounds <- list(x = c(-10, 40),
y = c(35, 70))
# Create a few points
cities <- data.frame(
name = c("Rome", "Berlin", "Paris"),
lon = c(12.5, 13.4, 2.35),
lat = c(41.9, 52.5, 48.85)
)
# Convert to sf POINTs
cities_sf <- st_as_sf(cities, coords = c("lon", "lat"), crs = 4326)
# Mapping Them
ggplot() +
geom_sf(data = world) +
geom_sf(data = cities_sf) +
coord_sf(xlim = europe_bounds$x,
ylim = europe_bounds$y) +
labs(title = "European Countries")
Complex Geoms
geom_sf Example
europe_bounds <- list(x = c(-10, 40),
y = c(35, 70))
# Create a few points
cities <- data.frame(
name = c("Rome", "Berlin", "Paris"),
lon = c(12.5, 13.4, 2.35),
lat = c(41.9, 52.5, 48.85)
)
# Convert to sf POINTs
cities_sf <- st_as_sf(cities, coords = c("lon", "lat"), crs = 4326)
# Create a line connecting Rome -> Berlin -> Paris
line_sf <- st_sfc(st_linestring(as.matrix(cities[, c("lon", "lat")])), crs = 4326)
# Mapping Them
ggplot() +
geom_sf(data = world) +
geom_sf(data = cities_sf) +
geom_sf(data = line_sf) +
coord_sf(xlim = europe_bounds$x,
ylim = europe_bounds$y) +
labs(title = "European Countries")
Exercises
Points vs. Lines
Use the dataset below:
- Make a scatterplot of turnout by year with
geom_point().
Exercises
Points vs. Lines
Use the dataset below:
- Now use
geom_line().
Exercises
Points vs. Lines
- Which geom makes more sense for these data? Why?
The line plot is usually more informative because it shows the trend in turnout over time.
Exercises
Bar vs. Column
Suppose you have survey data on respondents’ education:
- Plot the distribution of education levels with
geom_bar()
Exercises
Bar vs. Column
Suppose you have survey data on respondents’ education:
- Pre-count the values with
dplyr::count()and plot them withgeom_col():
Exercises
Boxplot vs. Histogram
We surveyed trust in government on a 1–10 scale:
- Plot the distribution with a histogram.
Exercises
Boxplot vs. Histogram
We surveyed trust in government on a 1–10 scale:
- Plot the same distribution with a boxplot.
Exercises
Boxplot vs. Histogram
- What information is easier to see in the histogram? In the boxplot?
- Easier to see the shape of the distribution (is it unimodal, bimodal, skewed?).
- You can spot clusters (e.g., many respondents around 5 and many around 7).
- You get a sense of frequency across different ranges.
- Easier to see the median and spread at a glance.
- Shows quartiles (middle 50% of the data).
- Highlights outliers explicitly.