Fun With Baby Names

(practice with the babynames data)

The babynames Package

Attach It

library(babynames)

This package exists in order to give us the data table babynames.

Study the Data Table

help(babynames)

babynames

# A tibble: 1,924,665 × 5
    year sex   name          n   prop
   <dbl> <chr> <chr>     <int>  <dbl>
 1  1880 F     Mary       7065 0.0724
 2  1880 F     Anna       2604 0.0267
 3  1880 F     Emma       2003 0.0205
 4  1880 F     Elizabeth  1939 0.0199
 5  1880 F     Minnie     1746 0.0179
 6  1880 F     Margaret   1578 0.0162
 7  1880 F     Ida        1472 0.0151
 8  1880 F     Alice      1414 0.0145
 9  1880 F     Bertha     1320 0.0135
10  1880 F     Sarah      1288 0.0132
# ℹ 1,924,655 more rows

Question

How popular has the name “Mary” been over the years?

babynames |> 
  filter(name == "Mary") |> 
  mutate(perc = prop * 100) |> 
  ggplot(aes(x = year, y = perc)) +
  geom_line(aes(color = sex)) +
  labs(x = NULL, y = "Percentage born named \"Mary\"")

Number vs. Percentage

babynames |> 
  filter(name == "Mary") |> 
  ggplot(aes(x = year, y = n)) +
  geom_line(aes(color = sex)) +
  labs(x = NULL, y = "Number born named \"Mary\"")

Try It!

How popular has your name been, over the years? (Work with percentages, not absolute number of births.)

Marking a Date

Popularity of Prince

Let’s investigate the popularity of the name “Prince” as a name for boys, since the year 1970.

We will make special note of 1978, the year that Prince released his classic album Purple Rain.

babynames |> 
  filter(name == "Prince" & year >= 1970 & sex == "M") |> 
  mutate(perc = prop * 100) |> 
  ggplot(aes(x = year, y = perc)) +
  geom_line() +
  geom_vline(aes(xintercept = 1978), color = "purple") +
  labs(
    x = NULL,
    y = 'Percentage of males named "Prince"',
    title = "There are more Princes, Now!",
    subtitle = "(after Purple Rain was released in 1978)"
  )

Two or More Names

Head-to-Head Matchup

Which name for males has been more popular over the years: “Homer”, or {insert unusual name of your choice}?

Some Ranking

Popular Female Names, by Decade

What are the top 5 most popular female names for each decade from the 1950s through the 2000-oughts?

To do this, we would need to find the decade in which each year occurs.

Getting the Decade

Try this, for various years:

Getting the Popular Female Names

tops <-
  babynames |> 
  # just 1950 through 2019, females:
  filter(year >= 1950 & year <= 2019 & sex == "F") |> 
  # make the decade column:
  mutate(
    decade = paste(
      floor(year / 10),
      "0s", sep = ""
    )
  ) |> 
  # get total babies for each name, in each decade:
  summarize(total = sum(n), .by = c(decade, name)) |> 
  # get top 5 names in each decade:
  slice_max(n = 5, order_by = total, by = decade) |> 
  arrange(decade, desc(total))

DT::datatable(tops, options = list(pageLength = 5))

What If …

… you had reversed the order of grouping?

bad_tops <-
  babynames |> 
  filter(year >= 1950 & year <= 2019 & sex == "F") |> 
  mutate(decade = paste(
    floor(year / 10),
    "0s", sep = ""
    )
  ) |> 
  # group by decades within each name, 
  # rather than names within each decade:
  summarize(total = sum(n), .by = c(name, decade)) |> 
  top_n(5, wt = total, by = decade) |> 
  arrange(decade, desc(total))