Data 502: Homework #4

library(magrittr)
library(tidyverse)

## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.1 ──

## ✓ ggplot2 3.3.5     ✓ purrr   0.3.4
## ✓ tibble  3.1.4     ✓ dplyr   1.0.7
## ✓ tidyr   1.1.3     ✓ stringr 1.4.0
## ✓ readr   2.0.1     ✓ forcats 0.5.1

## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## x tidyr::extract()   masks magrittr::extract()
## x dplyr::filter()    masks stats::filter()
## x dplyr::lag()       masks stats::lag()
## x purrr::set_names() masks magrittr::set_names()

library(ggplot2)

Babynames

# install.packages("babynames")
library(babynames)
?babynames

str(babynames)

## tibble [1,924,665 × 5] (S3: tbl_df/tbl/data.frame)
##  $ year: num [1:1924665] 1880 1880 1880 1880 1880 1880 1880 1880 1880 1880 ...
##  $ sex : chr [1:1924665] "F" "F" "F" "F" ...
##  $ name: chr [1:1924665] "Mary" "Anna" "Emma" "Elizabeth" ...
##  $ n   : int [1:1924665] 7065 2604 2003 1939 1746 1578 1472 1414 1320 1288 ...
##  $ prop: num [1:1924665] 0.0724 0.0267 0.0205 0.0199 0.0179 ...

Male Names

babynames_m <- babynames %>%
  filter(sex == "M")

str(babynames_m)

## tibble [786,372 × 5] (S3: tbl_df/tbl/data.frame)
##  $ year: num [1:786372] 1880 1880 1880 1880 1880 1880 1880 1880 1880 1880 ...
##  $ sex : chr [1:786372] "M" "M" "M" "M" ...
##  $ name: chr [1:786372] "John" "William" "James" "Charles" ...
##  $ n   : int [1:786372] 9655 9532 5927 5348 5126 3242 2632 2534 2444 2415 ...
##  $ prop: num [1:786372] 0.0815 0.0805 0.0501 0.0452 0.0433 ...

Editing down the information to reduce to ~top 25 male names for the last 50 years. To me, minimizing down is both more interesting/informative and easier to load quickly :)

babynames_m_19872017 <- babynames_m %>%
  filter(year > 1966)

babynames_m_19872017 <- babynames_m %>%
  filter(n > 28000)

I first thought that seeing progress over time is best viewed through a line chart. It’s interesting to see the partial lines where names pop in and out of the “top 25”. Althought I do believe the line graph is probably the most effective for something like this, it does look a little like “messy spaghetti”. I’ll likely edit down to less names in a future iteration.

ggplot(babynames_m_19872017, aes(x = year, y = n, color = name))+
  geom_line()

Just playing around with different geom types. The bar chart is actually a little interesting… More effective than I thought it would be, but probably not the best. It’s a little hard to see when certain names cross eachother, even with the alpha transparency usage. But very pretty! On a second look, it is actually really interesting to see how some of the names (although use more or less) are in a similar trend pattern… very interesting.

ggplot(babynames_m_19872017, aes(x = year, y = n, fill = name))+
  geom_col(alpha = 0.5)

Lets edit down a little more to look at the most popular names (top 10), but for the last 100 years. I’m curious to see what the longest standing boys’ name is on record… Even if it’s not number 1, could it arguably be the most popular name in history just out of sheer longevity?

babynames_m_19172017 <- babynames_m %>%
  filter(n > 51000)

ggplot(babynames_m_19172017, aes(x = year, y = n, color = name))+
  geom_line()+
  theme(rect = element_blank())+
  ggtitle("Top 10 Boys' Names for 100 Years")

Now it’s the Girls’ turn

Lets minimize the data to the most interesting

babynames_f <- babynames %>%
  filter(sex == "F")

summary(babynames_f)

##       year          sex                name                 n          
##  Min.   :1880   Length:1138293     Length:1138293     Min.   :    5.0  
##  1st Qu.:1954   Class :character   Class :character   1st Qu.:    7.0  
##  Median :1985   Mode  :character   Mode  :character   Median :   11.0  
##  Mean   :1975                                         Mean   :  151.4  
##  3rd Qu.:2003                                         3rd Qu.:   31.0  
##  Max.   :2017                                         Max.   :99686.0  
##       prop          
##  Min.   :2.360e-06  
##  1st Qu.:3.870e-06  
##  Median :7.180e-06  
##  Mean   :1.146e-04  
##  3rd Qu.:2.135e-05  
##  Max.   :7.238e-02

babynames_f_19862017 <- babynames_f %>%
  filter(year > 1986)

summary(babynames_f_19862017)

##       year          sex                name                 n          
##  Min.   :1987   Length:545743      Length:545743      Min.   :    5.0  
##  1st Qu.:1996   Class :character   Class :character   1st Qu.:    7.0  
##  Median :2003   Mode  :character   Mode  :character   Median :   11.0  
##  Mean   :2003                                         Mean   :  102.4  
##  3rd Qu.:2010                                         3rd Qu.:   28.0  
##  Max.   :2017                                         Max.   :55991.0  
##       prop          
##  Min.   :2.360e-06  
##  1st Qu.:3.310e-06  
##  Median :5.670e-06  
##  Mean   :5.189e-05  
##  3rd Qu.:1.419e-05  
##  Max.   :2.988e-02

babynames_f_19862017 <- babynames_f %>%
  filter(n > 28000)

Girls’ Top 30 Names for the Last 50 Years

ggplot(babynames_f_19862017, aes(x = year, y = n, color = name))+
  geom_line()

ggplot(babynames_f_19862017, aes(x = year, y = n, fill = name))+
  geom_col(alpha = 0.5)

…We’ll call it Top 11 for the girls. Interestingly you can see a big difference from this girls’ top 11 vs the boys’ top 10. Girl names have not lasted as long in popularity as boy names have. I guess girls are more interested in being fashionable? Haha, who knows.

babynames_f_19172017 <- babynames_f %>%
  filter(n > 45000)

ggplot(babynames_f_19172017, aes(x = year, y = n, color = name))+
  geom_line()+
  theme(rect = element_blank())+
  ggtitle("Top 11 Girls' Names for 100 Years")

Other fun visualizations

babynames_mf_top <- babynames %>%
  filter(n > 49000)

Popular names female vs male name trends.

Now seeing the top results combined you can see how boys are spread a little more evenly and girls have much more space and some pretty crazy outliers. I also manually messed with some of the coloring on these ones.

ggplot(babynames_mf_top, aes(x = year, y = n, color = sex))+
         geom_point(alpha = 0.7)+
  theme(plot.background = element_rect(fill = "black"),
        panel.background = element_rect(fill = "black"),
        axis.text = element_text(color = "cyan"),
        axis.line = element_blank(),
        panel.grid = element_blank(),
        legend.background = element_rect(fill = "black"),
        legend.text = element_text(color = "white"),
        legend.key = element_rect(fill = "black"))

Haha, super interesting what the lines do when there are so many points on the graph. It is definitely a different and interesting way to see density.

ggplot(babynames_mf_top, aes(x = year, y = n, color = sex))+
         geom_line(alpha = 0.7)+
  theme(plot.background = element_rect(fill = "black"),
        panel.background = element_rect(fill = "black"),
        axis.text = element_text(color = "cyan"),
        axis.line = element_blank(),
        panel.grid = element_blank(),
        legend.background = element_rect(fill = "black"),
        legend.text = element_text(color = "white"),
        legend.key = element_rect(fill = "black"))

I’m just messing around and having fun now like you said! :)

ggplot(babynames_mf_top, aes(x = year, y = n, fill = sex))+
  geom_area(alpha = 0.7)+
  scale_fill_manual(values = c("purple", "chartreuse"))+
  theme(plot.background = element_rect(fill = "black"),
        panel.background = element_rect(fill = "black"),
        axis.text = element_text(color = "white"),
        axis.title = element_blank(),
        axis.line = element_blank(),
        panel.grid = element_blank(),
        legend.background = element_rect(fill = "black"),
        legend.text = element_text(color = "white"),
        legend.key = element_rect(fill = "white"))

Oh, now this is interesting… the box plot is a surpisingly good way to look at the lasting power of a popularity of a name. I faceted the girls and boys charts to help clean up some of the dense clutter.

ggplot(babynames_mf_top) +
 aes(x = name, y = year, fill = name) +
 geom_boxplot() +
 scale_fill_hue() +
 theme_minimal() +
  theme(axis.line.x = element_blank(),
        axis.line.y = element_blank(),
        axis.text.x = element_text(angle=90, hjust=1),
        axis.title.y = element_blank())+
 facet_wrap(vars(sex), scales = "free")

Last… lets look at the number of times a babyname has been used (top female and male faceted in bar graphs) vs. how long they were popular.

ggplot(babynames_mf_top) +
    aes(x = name, fill = name, weight = n) +
    geom_bar() +
    scale_fill_hue(direction = 1) +
    labs(fill = "Name") +
    theme_minimal() +
    theme(
        plot.title = element_text(hjust = 0.5),
        axis.title.x = element_text(size = 7L),
        axis.text = element_text(color = "white"),
        axis.text.x = element_text(angle=90, hjust=1),
        panel.grid.major.x = element_blank(),
        panel.background = element_rect("black"),
        plot.background = element_rect("black"),
        legend.text = element_text(color = "white"),
        legend.title = element_text(color = "white")
    ) +
    facet_wrap(vars(sex), scales = "free")

This was fun! Thank you! - Rochelle Rafn