Specific names

The increased ethnic diversity can be tracked. Let’s look at the occurrence of Mohammed (and variations) in both countries over the timescale of the data

# Main variants
choice <- c("Mohammed","Mohammad","Mohamed","Mohamad")


uk_m <- uk %>% 
  filter(name %in% choice) %>% 
  group_by(year) %>% 
 summarize(count=sum(n,na.rm=TRUE)) %>% 
  plot_ly(x=~year,y=~count) %>% 
  layout(
    title="UK Births registered by name Mohammed (and variants) 1996-2015",
    xaxis=list(title=""),
    yaxis=list(title="")
  ) %>%  config(displayModeBar = F,showLink = F)
  

us_m <- us %>% 
  filter(name %in% choice) %>% 
  group_by(year) %>% 
 summarize(count=sum(n,na.rm=TRUE)) %>% 
  plot_ly(x=~year,y=~count) %>% 
  layout(
    title="UK Births registered by name Mohammed (and variants) 1880-2015",
    xaxis=list(title=""),
    yaxis=list(title="")
  ) %>%  config(displayModeBar = F,showLink = F)
  

tagList(
  uk_m,
  br(),
   us_m
)

Well that was a bit of a surprise - although admittedly I am not a UK resident. Is this down to less births to Moslems, more assimilation with use of traditional English names, concern about repercussions of using name..,?

Th absolute level in the US is much lower and took a big hit after 2001. It will be intersting to see how this pans out under the Trump presidency

Another area that merits research. Also. I may enhance this section to allow entry of any name

The Atlantic divide

What kind of correlation is there between the most popular names in the two countries

Here are four tables, based on 2015 data, which look at the top 10 male and female names in each location and how they rank across the ocean

#example code
topUKMales <- uk2015 %>% 
  filter(sex=="M") %>% 
  arrange(desc(n)) %>% 
  mutate(uk.rank=row_number()) %>% 
  slice(1:10)

a<- us2015 %>% 
  filter(sex=="M") %>% 
  arrange(desc(n)) %>% 
  mutate(us.rank=row_number()) %>% 
   right_join(topUKMales, by="name") %>% 
  select(`Top UK`=name,`uk rank`=uk.rank,`us rank`=us.rank) %>%
  DT::datatable(width=200,class='compact stripe hover row-border order-column',rownames=FALSE,options= list(paging = FALSE, searching = FALSE,info=FALSE))
 

All of the top 10 US choices, both female and male, make the top 100 in the UK but the reverse is not the case. perhaps because US culture hass more impact than vice-versa. This is particularly the case with 3rd ranked Harry (where there may be a royal component in the UK) and Poppy, 10th in the UK but not cracking the top 1000 in the States

Fashionability UK

The UK package also includes decade rankings of the top 95 names for both boys and girls throughout the last century. We can add the first decade of this century and the decade so far

# check what variables we need for update  
names(rankings)  
## [1] "name" "year" "rank" "sex"
  
# calculate and same for most recent two decades
  
uk2000 <-  uk %>% 
    filter(year>=2000&year<=2009) %>% 
    group_by(name,sex) %>% 
    summarize(count=sum(n)) %>% 
    arrange(desc(count)) %>% 
    ungroup() %>% 
   group_by(sex) %>% 
    mutate(rank=row_number(),year=2004) %>% 
    filter(rank<=95) %>% 
    select(-count)

uk2010 <-  uk %>% 
    filter(year>=2010&year<=2015) %>% 
    group_by(name,sex) %>% 
    summarize(count=sum(n)) %>% 
    arrange(desc(count)) %>% 
    ungroup() %>% 
   group_by(sex) %>% 
    mutate(rank=row_number(),year=2014) %>% 
    filter(rank<=95) %>% 
    select(-count)

rankingsUK <- bind_rows(rankings,uk2000,uk2010)

## look at those appearing in Top Ten most decades


rankingsUK %>%
  filter(sex=="M"&rank<11) %>%
  count(name, sort=TRUE)
## # A tibble: 41 x 2
##       name     n
##      <chr> <int>
##  1   James    10
##  2    John     7
##  3 William     7
##  4   David     6
##  5 Michael     6
##  6  Thomas     6
##  7  George     5
##  8  Robert     5
##  9    Paul     4
## 10 Richard     4
## # ... with 31 more rows


rankingsUK %>%
  filter(sex=="F"&rank<11) %>%
  count(name, sort=TRUE)
## # A tibble: 71 x 2
##         name     n
##        <chr> <int>
##  1  Margaret     6
##  2      Mary     6
##  3 Elizabeth     4
##  4     Doris     3
##  5   Dorothy     3
##  6     Emily     3
##  7      Emma     3
##  8   Jessica     3
##  9  Patricia     3
## 10    Sophie     3
## # ... with 61 more rows

No name placed in the top 10 for each of the twelve decades with James, at ten, the most consistently popular. Margaret and Mary seem pretty dated now

Let’s look at the top 10 for the first and last decade and see how they have placed over the years

#example code
oldMales <- rankingsUK %>% 
  filter(sex=="M"&rank<11&year==1904) %>% 
  pull(name)


newMales <-rankingsUK %>% 
  filter(sex=="M"&rank<11&year==2014) %>% 
  pull(name)

allMales <- base::union(oldMales,newMales) # 18 so slight ovelap


e <-rankingsUK %>% 
  filter(name %in% allMales) %>% 
  plot_ly(x=~year,y=~rank,color=~name,
          hoverinfo="text",
          text=~paste0(name,"<br>rank: ",rank)) %>% 
  add_markers() %>% 
  add_lines(showlegend=FALSE) %>% 
  layout(
    yaxis=list(autorange="reversed"),
    xaxis=list(title="decade")
  )

I have included lines to easier see trends over time but this can be misleading e.g Matthew does not appear during 1920-1950 descades The points reflect the actual ranking

Although none of the girl names appear in the top ten’s a century apart, there have been some current favourites e.g Emily and Lily that were relatively popular in Edwardian times before going out of fashion for quite a while.

John used to be a perennial favourite - and indeed for centuries past. In 2015, it was ranked 108th

Number Ones in the USA

A recent chart on No 1 tennis players over history pompted me to look at the equivalent layout for girl and boy names

names(us)
## [1] "year" "sex"  "name" "n"    "prop"

usRanking_male <- us %>% 
  filter(sex=="M") %>% 
  arrange(desc(n)) %>% 
  group_by(year) %>% 
  mutate(rank=row_number())

usMaleOnes <- usRanking_male %>% 
  filter(rank==1) %>% 
  pull(name) %>% 
  unique()
  

nameArrangeM <- 
  usRanking_male %>%  
  arrange(year) %>% 
    filter(name %in% usMaleOnes&rank==1) %>% 
   group_by(name) %>%
  slice(1) %>% 
  arrange(year) %>% 
  pull(name)

nameArrangeM
## [1] "John"    "Robert"  "James"   "Michael" "David"   "Jacob"   "Noah"


usRanking_male$nameFactor  <- factor(usRanking_male$name, levels = c("Noah","Jacob","David","Michael","James","Robert","John"))



g <-usRanking_male %>%  
  arrange(year) %>% # needed to ensure works
  filter(name %in% usMaleOnes) %>% 
  mutate(op=sqrt(1/rank)) %>% 
  plot_ly(x=~year,y=~nameFactor, height=250,
          hoverinfo="text",
          text=~paste0(year,"<br>Rank: ",rank)) %>% 
  add_markers(marker = list(opacity = ~op)) %>% 
  layout(title="US Male Babynames which have ever held Annual Top spot<br>Hover for details",
         xaxis=list(title=""),
         yaxis=list(title="")
  )

Noah has made quite the leap in the past 25 years. Jennifer did not even register before 1916 but was number 1 in 1970 before suffering a pretty precipitous decline

There has been quite a bit of work based on the such as estimating how old you are from your name and the use of names that can be used by either gender and it might be interesting to extend those analyses to the UK (or for that matter other countries, if made available) data