Project 2 - Dataset #2

I am using the below dataset, “Baby Names from Social Security Card Applications - National Level Data”

https://catalog.data.gov/dataset/baby-names-from-social-security-card-applications-national-level-data

Data for each year is stored in a different text file. I am only analyzing data from 2000-2017 but this logic will handle all the files in the folder.

Load Package

library(tidyverse)
require(stringr)
library(kableExtra)

Load data

col_names <- c('name', 'sex', 'count', 'year' )

# This is used to store all  names from all  files. 
Names <- data.frame()


# This funciton will read the file and append names to "Names" varible outside the function. 
read_files <- function(file_name){
    
  x <- read.csv(file_name, header = F, stringsAsFactors =  F) %>%  
      mutate(year = as.numeric( str_extract(file_name,'\\d+')))
  
  Names <<-  bind_rows(x,Names)
  
  return('')  
}

# Read names from all the files with "yob" in the file name.
filenames <- list.files( pattern="yob", full.names=TRUE)
jx <- lapply(filenames, FUN = read_files)

colnames(Names)<- col_names

boys_names <- Names %>%
  filter(sex =='M') 


girls_names <- Names %>%
  filter(sex =='F') 

2017 Top 10 Names

popular_names_17_boys <- boys_names %>% 
            filter(year == 2017) %>%
            top_n(10, count) %>%
            select(name, count,sex)

popular_names_17_girls <- girls_names %>% 
  filter(year == 2017) %>%
  top_n(10,count) %>%
  select(name, count,sex) 

top_names_17 <- data.frame(popular_names_17_girls$name, popular_names_17_boys$name)
colnames(top_names_17) <- c("Girls","Boys")

  kable(top_names_17) %>%
    kable_styling(bootstrap_options = c("striped", "hover", "condensed", "responsive"))

Girls	Boys
Emma	Liam
Olivia	Noah
Ava	William
Isabella	James
Sophia	Logan
Mia	Benjamin
Charlotte	Mason
Amelia	Elijah
Evelyn	Oliver
Abigail	Jacob

2017 popular names over the years.

  These plots show the popularity of the names over the last 17 years. Names like Jacob are becoming less popular while Liam and other names becoming more popular. This is also happening with girl's name.    

popular_names_17_boys_trend <- Names %>% inner_join(popular_names_17_boys, by =c ('name','sex'))
popular_names_17_girls_trend <- Names %>% inner_join(popular_names_17_girls, by =c ('name','sex'))

popular_names_17_boys_trend <- Names %>% inner_join(popular_names_17_boys, by =c ('name','sex'))
popular_names_17_girls_trend <- Names %>% inner_join(popular_names_17_girls, by =c ('name','sex'))

ggplot(popular_names_17_boys_trend, aes(x= year, y= count.x, color = name)) + 
  geom_line()+
geom_text(data = popular_names_17_boys_trend %>% filter(year == last(year)), aes(label = name, 
                                                                                                x = year + 0.5, 
                                                                                                y = count.x, 
                                                                                                color = name)) 

ggplot(popular_names_17_girls_trend, aes(x= year, y= count.x, color = name)) + 
  geom_line()+
geom_text(data = popular_names_17_girls_trend %>% filter(year == first(year)), aes(label = name, 
                                                                       x = year + 0.5, 
                                                                       y = count.x, 
                                                                       color = name)) 

Top 5 names per year over the years

Looking at the top 5 names for boys and girls per year from 2000 to 2017 shows the same patter as above.  

# Top five names over the years
top_five_boys <- boys_names %>%
  group_by(year) %>%
  top_n(5,count) %>%
  select(name)

## Adding missing grouping variables: `year`

# Get the distinct top names over the years. For example, Jacob was a popular name for many years.
top_boy_names <-data.frame(top_five_boys$name) %>% distinct()
colnames(top_boy_names)<-c('name')
top_five_boys_per <- boys_names %>% inner_join(top_boy_names, by =c ('name'))

## Warning: Column `name` joining character vector and factor, coercing into
## character vector

top_five_girls <- girls_names %>%
  group_by(year) %>%
  top_n(5,count) %>%
  select(name)

## Adding missing grouping variables: `year`

top_girl_names <-data.frame(top_five_girls$name) %>% distinct()
colnames(top_girl_names)<-c('name')
top_five_grils_per <- girls_names %>% inner_join(top_girl_names, by =c ('name'))

## Warning: Column `name` joining character vector and factor, coercing into
## character vector

top_names<- c(top_girl_names, top_boy_names)

  kable(top_names) %>%
    kable_styling(bootstrap_options = c("striped", "hover", "condensed", "responsive"))

x Emma Olivia Ava Isabella Sophia Emily Madison Abigail Hannah Alexis Ashley Sarah

x Liam Noah William James Logan Mason Jacob Ethan Jayden Michael Alexander Joshua Daniel Matthew Andrew Christopher

ggplot(top_five_boys_per, aes(x= year, y = count, color = name)) + 
  geom_line() +
geom_text(data = top_five_boys_per %>% filter(year == last(year)), aes(label = name, 
                                                                                 x = year + 0.5, 
                                                                                 y = count, 
                                                                                 color = name)) 

ggplot(top_five_grils_per, aes(x= year, y= count, color = name)) + 
  geom_line()+
geom_text(data = top_five_grils_per %>% filter(year == first(year)), aes(label = name, 
                                                                                   x = year + 0.5, 
                                                                                   y = count, 
                                                                                   color = name)) 

Popular names like Jacob, Michael, Matthew and Joshua from early 2000’s are loosing popularity while less popular names like Jayden, Liam, Mason and Logan are gaining popularity or are even more popular. For example, Jacob really popular in early 2000s while Liam was one of the least popular names. In 2015, Liam was more popular than Jacob. Same pattern is happening for girls names too.