assignment 4-infections

# load libraries
library(ggplot2)
library(readr)
library(dplyr)

# reading the space-separated txt file
infections <- read_table("/Users/gretacapelletti/Downloads/infections1.txt")

── Column specification ─────────────────────────────────────────────────────────────────────
cols(
  infections = col_double(),
  ufo2010 = col_double(),
  pop = col_double()
)

# quick look at the data
head(infections)

NA

ggplot(infections, aes(x = infections)) +
  geom_histogram(binwidth = 500, fill = "lightblue", color = "black") +
  labs(title = "Histogram of Infections", x = "Infections", y = "Count")

# most of the values are under 1000, so the majority of cases are low
# a few bars on the right show really high infection numbers — definitely outliers
# overall, the distribution is skewed to the right (lots of low values, some big ones)

# add index to use as x-axis
infections$index <- 1:nrow(infections)

ggplot(infections, aes(x = index, y = infections)) +
  geom_line(color = "red") +
  labs(title = "Infection Trend", x = "Index", y = "Infections")

# infection numbers go up and down a lot
# there are a few big spikes, maybe outliers or hot spots
# nothing super steady — it jumps randomly

ggplot(infections, aes(x = ufo2010, y = infections)) +
  geom_point(color = "blue") +
  labs(title = "Infections vs UFO Sightings", x = "UFO Sightings", y = "Infections")

# no clear pattern, but fun to see if there's any connection

# there's no clear relationship between UFO sightings and infections
# most of the points are low for both, with a few random outliers
# doesn’t really show a trend, but fun to explore

ggplot(infections, aes(x = pop, y = infections)) +
  geom_point(color = "purple") +
  scale_x_log10() +
  scale_y_log10() +
  labs(title = "Infections vs Population (log scale)", x = "Population (log)", y = "Infections (log)")

# log scale makes it easier to see patterns in messy data

# The original scatter plot was hard to interpret because one or two really large population values stretched the x-axis.
# Most of the other data points were clustered in one tiny area, making it difficult to spot any trends.
# To fix this, I applied a log scale to both the x-axis (population) and y-axis (infections).
# This transformation doesn’t change the data — it just adjusts how it’s displayed.
# The log scale makes it easier to visualize the relationship between infections and population,
# especially when the data has extreme values or is spread out across a wide range.
# After applying the log scale, we can better see that infections tend to increase as population increases.

ggplot(infections, aes(x = pop, y = infections)) +
  geom_point(color = "purple") +
  geom_smooth(method = "lm", se = FALSE, color = "black") +
  scale_x_log10() +
  scale_y_log10() +
  labs(title = "Infections vs Population (Log Scale + Trend Line)", x = "Population (log)", y = "Infections (log)")

# This version shows the relationship between infections and population using log scales.
# Log scale helps adjust for the wide range in values — especially the large outliers.
# The regression line makes it clear that as population increases, infection numbers also tend to rise.
# This visualization is more balanced and easier to interpret than the original scatter plot.