Homework 3 — Dates, Factors + EDA
Dev Narang
This homework has two parts. Part 1 uses lubridate and
factors on NYC flight data. Part 2 does a full EDA on the built-in
airquality dataset.
Part 1 — NYC Flights: Dates and Factors
# install.packages(c("nycflights13", "lubridate", "zoo", "forcats")) # if needed
library(nycflights13)
library(dplyr)##
## Attaching package: 'dplyr'## The following objects are masked from 'package:stats':
##
## filter, lag## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, unionlibrary(lubridate)##
## Attaching package: 'lubridate'## The following objects are masked from 'package:base':
##
## date, intersect, setdiff, unionlibrary(zoo)##
## Attaching package: 'zoo'## The following objects are masked from 'package:base':
##
## as.Date, as.Date.numericlibrary(forcats)
data(flights)Quick look:
str(flights)## tibble [336,776 × 19] (S3: tbl_df/tbl/data.frame)
## $ year : int [1:336776] 2013 2013 2013 2013 2013 2013 2013 2013 2013 2013 ...
## $ month : int [1:336776] 1 1 1 1 1 1 1 1 1 1 ...
## $ day : int [1:336776] 1 1 1 1 1 1 1 1 1 1 ...
## $ dep_time : int [1:336776] 517 533 542 544 554 554 555 557 557 558 ...
## $ sched_dep_time: int [1:336776] 515 529 540 545 600 558 600 600 600 600 ...
## $ dep_delay : num [1:336776] 2 4 2 -1 -6 -4 -5 -3 -3 -2 ...
## $ arr_time : int [1:336776] 830 850 923 1004 812 740 913 709 838 753 ...
## $ sched_arr_time: int [1:336776] 819 830 850 1022 837 728 854 723 846 745 ...
## $ arr_delay : num [1:336776] 11 20 33 -18 -25 12 19 -14 -8 8 ...
## $ carrier : chr [1:336776] "UA" "UA" "AA" "B6" ...
## $ flight : int [1:336776] 1545 1714 1141 725 461 1696 507 5708 79 301 ...
## $ tailnum : chr [1:336776] "N14228" "N24211" "N619AA" "N804JB" ...
## $ origin : chr [1:336776] "EWR" "LGA" "JFK" "JFK" ...
## $ dest : chr [1:336776] "IAH" "IAH" "MIA" "BQN" ...
## $ air_time : num [1:336776] 227 227 160 183 116 150 158 53 140 138 ...
## $ distance : num [1:336776] 1400 1416 1089 1576 762 ...
## $ hour : num [1:336776] 5 5 5 5 6 5 6 6 6 6 ...
## $ minute : num [1:336776] 15 29 40 45 0 58 0 0 0 0 ...
## $ time_hour : POSIXct[1:336776], format: "2013-01-01 05:00:00" "2013-01-01 05:00:00" ...head(flights)## # A tibble: 6 × 19
## year month day dep_time sched_dep_time dep_delay arr_time sched_arr_time
## <int> <int> <int> <int> <int> <dbl> <int> <int>
## 1 2013 1 1 517 515 2 830 819
## 2 2013 1 1 533 529 4 850 830
## 3 2013 1 1 542 540 2 923 850
## 4 2013 1 1 544 545 -1 1004 1022
## 5 2013 1 1 554 600 -6 812 837
## 6 2013 1 1 554 558 -4 740 728
## # ℹ 11 more variables: arr_delay <dbl>, carrier <chr>, flight <int>,
## # tailnum <chr>, origin <chr>, dest <chr>, air_time <dbl>, distance <dbl>,
## # hour <dbl>, minute <dbl>, time_hour <dttm># Q1. Create a column dep_datetime by combining year, month, day, and dep_time into a
# POSIXct datetime using lubridate's make_datetime().
# Hint: hour = dep_time %/% 100, minute = dep_time %% 100
# Show the first 5 rows with year, month, day, dep_time, and dep_datetime.
flights <- flights |>
mutate(dep_datetime = make_datetime(year, month, day,
dep_time %/% 100, dep_time %% 100))
flights |>
select(year, month, day, dep_time, dep_datetime) |>
head(5)## # A tibble: 5 × 5
## year month day dep_time dep_datetime
## <int> <int> <int> <int> <dttm>
## 1 2013 1 1 517 2013-01-01 05:17:00
## 2 2013 1 1 533 2013-01-01 05:33:00
## 3 2013 1 1 542 2013-01-01 05:42:00
## 4 2013 1 1 544 2013-01-01 05:44:00
## 5 2013 1 1 554 2013-01-01 05:54:00# Q2. After creating dep_datetime, use lubridate's month() to filter flights that
# departed in JUNE 2013. How many flights are there?
# (Hint: filter(month(dep_datetime) == 6))
june_flights <- flights |>
filter(month(dep_datetime) == 6)
nrow(june_flights)## [1] 27234# Q3. The carrier column is a character. Convert it to a factor and check its levels.
flights <- flights |>
mutate(carrier = factor(carrier))
levels(flights$carrier)## [1] "9E" "AA" "AS" "B6" "DL" "EV" "F9" "FL" "HA" "MQ" "OO" "UA" "US" "VX" "WN"
## [16] "YV"# Q4. Use fct_collapse() to keep "UA", "AA", and "DL" as their own levels and lump
# everything else into "Other". Then count flights per recoded carrier level.
# (Hint: see the fct_collapse demo from the Wrangling Activity Part H)
flights <- flights |>
mutate(carrier_grouped = fct_collapse(carrier,
UA = "UA", AA = "AA", DL = "DL",
other_level = "Other"))
flights |>
count(carrier_grouped)## # A tibble: 4 × 2
## carrier_grouped n
## <fct> <int>
## 1 AA 32729
## 2 DL 48110
## 3 UA 58665
## 4 Other 197272# Q5. Missing data: how many flights have NA for dep_delay? Filter them out and report
# the remaining row count.
sum(is.na(flights$dep_delay))## [1] 8255flights_clean <- flights |>
filter(!is.na(dep_delay))
nrow(flights_clean)## [1] 328521Part 2 — Airquality EDA
library(ggplot2)
data("airquality")# Q6. For Ozone, Temp, and Wind: compute mean, median, sd, min, max
# (use na.rm = TRUE where needed).
vars <- c("Ozone", "Temp", "Wind")
sapply(airquality[vars], function(x) c(
mean = mean(x, na.rm = TRUE),
median = median(x, na.rm = TRUE),
sd = sd(x, na.rm = TRUE),
min = min(x, na.rm = TRUE),
max = max(x, na.rm = TRUE)
))## Ozone Temp Wind
## mean 42.12931 77.88235 9.957516
## median 31.50000 79.00000 9.700000
## sd 32.98788 9.46527 3.523001
## min 1.00000 56.00000 1.700000
## max 168.00000 97.00000 20.700000Q7. Compare the mean and median for each variable. What does the relationship between mean and median suggest about distribution skewness? What does the standard deviation tell you about variability?
For Ozone, the mean (42.13) is well above the median (31.5), indicating a right-skewed distribution (a few high values pull the mean up). For Temp (mean 77.88, median 79) and Wind (mean 9.96, median 9.70), the mean and median are very close, so both are roughly symmetric. The standard deviations show variability: Ozone’s SD (≈33) is huge relative to its mean, so it’s highly variable, while Temp (SD ≈9.5) and especially Wind (SD ≈3.5) are much more stable.
# Q8. Make a histogram of Ozone.
ggplot(airquality, aes(x = Ozone)) +
geom_histogram(binwidth = 10, fill = "steelblue", color = "white") +
labs(title = "Distribution of Ozone", x = "Ozone (ppb)", y = "Count")## Warning: Removed 37 rows containing non-finite outside the scale range
## (`stat_bin()`).
Q9. Describe the shape of the Ozone distribution. Any outliers or unusual features?
The Ozone distribution is right-skewed (unimodal with a long right tail). Most readings fall between roughly 0 and 50 ppb, but there are a handful of high values stretching up to ~168 ppb, which act as outliers on the high end.
# Q10. Create a new column month_name with full month names (May–September) using case_when.
# Then make a boxplot of Ozone by month_name.
# (Hint: case_when was covered in the Wrangling Activity.)
airquality <- airquality |>
mutate(month_name = case_when(
Month == 5 ~ "May",
Month == 6 ~ "June",
Month == 7 ~ "July",
Month == 8 ~ "August",
Month == 9 ~ "September"
),
month_name = factor(month_name,
levels = c("May", "June", "July", "August", "September")))
ggplot(airquality, aes(x = month_name, y = Ozone)) +
geom_boxplot(fill = "lightblue") +
labs(title = "Ozone by Month", x = "Month", y = "Ozone (ppb)")## Warning: Removed 37 rows containing non-finite outside the scale range
## (`stat_boxplot()`).
Q11. Which month has the highest median Ozone? Are there outliers in any month?
July has the highest median Ozone (≈60 ppb), with August second (≈52). The summer months (July, August, September) show high outliers in their boxplots, while May and June are lower and tighter.
# Q12. Scatterplot of Temp vs Ozone, colored by Month.
ggplot(airquality, aes(x = Temp, y = Ozone, color = factor(Month))) +
geom_point(size = 2) +
labs(title = "Temp vs Ozone", x = "Temp (F)", y = "Ozone (ppb)", color = "Month")## Warning: Removed 37 rows containing missing values or values outside the scale range
## (`geom_point()`).
Q13. Is there a visible relationship between temperature and ozone?
Yes — there is a clear positive relationship: as temperature increases, Ozone tends to increase. Higher ozone values cluster at higher temperatures.
# Q14. Compute the correlation matrix for Ozone, Temp, and Wind.
# (Hint: cor(airquality[, c("Ozone","Temp","Wind")], use = "complete.obs"))
cor(airquality[, c("Ozone", "Temp", "Wind")], use = "complete.obs")## Ozone Temp Wind
## Ozone 1.0000000 0.6983603 -0.6015465
## Temp 0.6983603 1.0000000 -0.5110750
## Wind -0.6015465 -0.5110750 1.0000000Q15. Which pair has the strongest correlation? What does that suggest?
Ozone and Temp have the strongest correlation (r ≈ 0.70, positive). This suggests warmer temperatures are associated with higher ozone levels. Ozone and Wind are negatively correlated (r ≈ -0.60), meaning higher winds tend to disperse ozone and lower it.
# Q16. Generate a summary table grouped by Month with: count, mean Ozone, mean Temp,
# mean Wind for each month.
airquality |>
group_by(Month) |>
summarize(
count = n(),
mean_ozone = mean(Ozone, na.rm = TRUE),
mean_temp = mean(Temp, na.rm = TRUE),
mean_wind = mean(Wind, na.rm = TRUE)
)## # A tibble: 5 × 5
## Month count mean_ozone mean_temp mean_wind
## <int> <int> <dbl> <dbl> <dbl>
## 1 5 31 23.6 65.5 11.6
## 2 6 30 29.4 79.1 10.3
## 3 7 31 59.1 83.9 8.94
## 4 8 31 60.0 84.0 8.79
## 5 9 30 31.4 76.9 10.2Q17. Which month has the highest average ozone? How do temperature and wind vary across months?
August has the highest average Ozone (≈60 ppb), with July a close second (≈59). Temperature follows the same pattern, peaking in July–August, while wind is generally lower during those warm summer months — consistent with the positive Ozone–Temp and negative Ozone–Wind correlations.