Tidy Tuesday 2021 Week 41

TidyTuesday week 41: Registered Nurses, data from Data.World.

library(tidyverse)
library(janitor)
library(ggtext)
library(scales)
library(geofacet)
library(colorspace)
library(biscale)
library(cowplot)

nurses <- readr::read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2021/2021-10-05/nurses.csv') %>% clean_names()

Rows: 1242 Columns: 22
── Column specification ───────────────────────────────────────────────────────────────────────────────────────────────────
Delimiter: ","
chr  (1): State
dbl (21): Year, Total Employed RN, Employed Standard Error (%), Hourly Wage Avg, Hourly Wage Median, Annual Salary Avg,...

ℹ Use `spec()` to retrieve the full column specification for this data.
ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.

range(nurses$year)

[1] 1998 2020

Tile map

• shared on Twitter

nurses$st <- state.abb[match(nurses$state, state.name)]

df20 = nurses %>% filter(year==2020) %>% 
  mutate(st=ifelse(state=="District of Columbia","DC",st)) %>%
  mutate(st=ifelse(state=="Puerto Rico","PR",st)) %>% 
  drop_na(st) 

df1 = df20 %>%
  select(state, st, year, total_employed_rn, hourly_wage_median) %>%
  bi_class(x=hourly_wage_median, y=total_employed_rn, style="quantile",dim=3) 

create_gradient_state_tile_map <- function(state, value, title, subtitle, caption, legend_title, state_grid='us_state_with_DC_PR_grid2') {
  
  df <- as.tibble(data.frame(state, value))
  
  fig <- df %>% 
    mutate(x = 1) %>% 
    mutate(label_y = .5) %>%  
    mutate(label_x = 1) %>% 
    ggplot()+
    geom_bar(mapping=aes(x=x, fill=value), width=.4)  +
    facet_geo(~ state, grid=state_grid) +
    labs(title=title, subtitle=subtitle, caption=caption) +
    geom_text(aes(x=label_x, y=label_y, label=state, color=value),size=3, show.legend=F, family="sans") 
  
  return(fig)
}

# excl. Guam and Virgin Islands
p1 = create_gradient_state_tile_map(df1$st, df1$bi_class, 
                                    title='US-based Registered Nurses Employment and Wage in 2020', legend_title = "",
                                    subtitle="<span style = 'color:#012a4a;'><b>Total employed registered nurses</b></span> and <span style = 'color:#012a4a;'><b>median hourly wage</b></span>, by US state\n",
                               caption="Note: Data from Guam and Virgin Islands are not presented<br>#TidyTuesday Week 41 | Data from Data.World") + 
  bi_scale_fill(pal="DkCyan",dim=3, guide="none") +
  scale_color_manual(values=c("grey10","grey10","white","white","white","white","white","white","white")) +
  theme_void(base_size=10, base_family = "sans") + 
  theme(strip.text.x = element_blank(),
        plot.margin = unit(c(.5,4,.5,2), "cm"),
        plot.title=element_text(size=14, face="bold", color="#012a4a"),
        plot.subtitle=element_markdown(size=8, color="#011c31", margin=margin(t=5,b=18)),
        legend.title=element_text(size=9),
        plot.caption = element_markdown(size=5.6, color="#011c31",margin=margin(t=30), lineheight=1.5, hjust=0)) + 
  guides(fill = guide_colorbar(title="Count",
                              title.position = "top", 
                              barwidth = unit(.5, "lines"), 
                              barheight = unit(10, "lines"))) 

p2 = bi_legend(pal = "DkCyan", 
            dim = 3,
            ylab = "Total employed",
            xlab = "Median hourly wage",
            size = 2.5) + 
  theme(panel.border = element_blank(),
        axis.text = element_blank(),
        axis.title.x = element_text(size = 6, family="sans",
                                    color = "#011c31", margin=margin(t=-5)),
        axis.title.y = element_text(size = 6, family="sans",
                                    color = "#011c31", margin=margin(r=-5)),
        legend.text = element_text(size = 6),
        plot.background = element_blank(),
        legend.text.align = 0)

ggdraw() +
  draw_plot(p1, 0, 0, 1, 1) +
  draw_plot(p2, 0.72, 0.04, 0.25, 0.25) 

AlT text: Bivariate heatmap of US-based registered nurses’ median hourly rate and total employment by state (excluding Guam and Virgin Island). The visualization shows that in comparison to other states: Utah, South Dakota, West Virginia and Puerto Rico have the least total registered nurses employment and lowest median hourly rate. California, Minnesota, Massachusetts, New York, and New Jersey have the most total employment and highest median hourly rate compared to the other states.

Tile map v2

• add labels for median hourly rage and total employment

# bivariate palette reference: https://nowosad.github.io/post/cbc-bp2/
library(pals)
#brewer.seqseq2()
bi_pal <- bi_pal_manual(val_1_1 = "#f3f3f3", val_1_2 = "#b4d3e1", val_1_3 = "#509dc2",
                        val_2_1 = "#f3e6b3", val_2_2 = "#b3b3b3",val_2_3 = "#376387",
                        val_3_1 = "#f3b300", val_3_2 = "#b36600",val_3_3 = "#000000")
# label colors
df2 = df1 %>% 
  mutate(col=case_when(str_detect(bi_class,"3")~"white",TRUE~"black")) %>%
  mutate(col=case_when(bi_class=="3-1"~"black",TRUE~col))

p1b = df2 %>%
  ggplot() +
  geom_rect(aes(fill=bi_class), xmin=-1, xmax=1, ymin=-1, ymax=1, color="white", show.legend = F) +
  geom_richtext(aes(color=col, label=glue::glue("<span style='font-size: 11px;'>{st}</span><br>{dollar(hourly_wage_median)}<br>{round(total_employed_rn/1000,1)}K")),
                x=0.5, y=0.45, size=1.5, hjust=.5, fill=NA, label.color=NA, show.legend = F) +
  scale_color_identity() +
  facet_geo(vars(st), grid=us_state_with_DC_PR_grid2) +
  bi_scale_fill(pal=bi_pal, dim=3, guide="none") + 
  theme_void(base_size = 9) +
  theme(strip.text = element_blank(),
        plot.margin = unit(c(.5, 3.2, .5, 2.5), "cm"),
        plot.title=element_markdown(face="bold"),
        plot.subtitle=element_markdown(margin=margin(t=4, b=15)),
        plot.caption=element_text(size=6, hjust=0, margin=margin(b=0, t=30))) +
  labs(title="2020 US-based Registered Nurses Wage and Employment",
       subtitle="Median hourly wage and total employed, by state",
       caption="Note: Data from Guam and Virgin Islands are not presented\nData source: Data.World")

p2b = bi_legend(pal =  bi_pal,
            dim = 3,
            ylab = "Total employed",
            xlab = "Median hourly wage",
            size = 2.5) + 
  theme(panel.border = element_blank(),
        axis.text = element_blank(),
        axis.title.x = element_text(size = 6, family="sans",
                                    color = "#011c31", margin=margin(t=-5)),
        axis.title.y = element_text(size = 6, family="sans",
                                    color = "#011c31", margin=margin(r=-5)),
        legend.text = element_text(size = 6),
        plot.background = element_blank(),
        legend.text.align = 0)

ggdraw() +
  draw_plot(p1b, 0, 0, 1, 1) +
  draw_plot(p2b, 0.645, 0.04, 0.25, 0.25) 

Western region table

library(gt)
library(gtExtras)
library(usmap)
library(patchwork)

nurses$st = state.abb[match(nurses$state, state.name)]

tab1 = df20 %>%
  filter(st %in% .west_region) %>%
  mutate(division= case_when(st %in% .mountain ~ "Mountain division",
                             st %in% .pacific ~ "Pacific division")) %>%
  #group_by(state) %>% 
  #mutate(ratio=total_employed_rn/total_employed_healthcare_state_aggregate) %>%
  #ungroup() %>%
  select(state,division,  annual_salary_avg, hourly_wage_avg,wage_salary_standard_error_percent,
         total_employed_rn,total_employed_healthcare_state_aggregate,yearly_total_employed_state_aggregate,
         location_quotient) %>%
  arrange(desc(annual_salary_avg))

tab1 %>% 
  gt(groupname_col = "division") %>%
  tab_header(title = "WESTERN U.S. REGISTERED NURSES 2020") %>%
  tab_source_note(source_note="Data source: Data.World") %>%
  cols_label(
    annual_salary_avg = html("Annual ($)"),
    hourly_wage_avg = html("Hourly ($)"),
    wage_salary_standard_error_percent = html("SE (%)"),
    total_employed_rn = html("RN"),
    total_employed_healthcare_state_aggregate=html("Healthcare"),
    yearly_total_employed_state_aggregate = html("Yearly"),
    location_quotient=html("Quotient"),
    state=html("State"),
  ) %>%
  tab_spanner(
    label = "Total Employed",
    columns = total_employed_rn:yearly_total_employed_state_aggregate
  ) %>%
  tab_spanner(
    label = "Salary/Wage",
    columns = annual_salary_avg:wage_salary_standard_error_percent
  ) %>%
  tab_spanner(
    label = "Location",
    columns = location_quotient
  ) %>%
  tab_options(table.font.size = "14px",
              heading.title.font.size = "18px",column_labels.border.bottom.color = "grey",
              data_row.padding = px(4),
              heading.padding = px(10),
              ) %>%
  gt_hulk_col_numeric(location_quotient) %>%
  #gt_hulk_col_numeric(hourly_wage_avg) %>%
  gt_hulk_col_numeric(annual_salary_avg) %>%
  gt_hulk_col_numeric(total_employed_rn) %>%
  cols_align(
    align = "center",
    columns = annual_salary_avg:location_quotient
  ) %>%
  cols_width(annual_salary_avg:location_quotient ~ px(80)) %>%
  cols_width(state ~ px(110)) %>%
  fmt_number(annual_salary_avg:location_quotient, use_seps = T, drop_trailing_zeros = T) %>%
  tab_footnote(
    footnote = md("Total employed registered nurses"),
    locations = cells_column_labels(columns = total_employed_rn)
  ) %>%
  tab_footnote(
    footnote = md("Total employed healthcare, state aggregate"),
    locations = cells_column_labels(columns = total_employed_healthcare_state_aggregate)
  ) %>%
  tab_footnote(
    footnote = md("Yearly total employed, state aggregte"),
    locations = cells_column_labels(columns = yearly_total_employed_state_aggregate)
  ) %>%
  tab_style(
    style = list(
      cell_text(font=google_font(
        name = "Libre Franklin"), weight='800',align = "left",color='#203B46')),
    locations = cells_title(groups = "title")
  ) %>%
  tab_style(style = cell_text(font = google_font("Source Sans Pro"), 
            weight = 400), locations = cells_body()) %>%
  tab_style(style = cell_text(size = px(12.5), color = "grey30", 
        font = google_font("Source Sans Pro"), transform = "uppercase"), 
        locations = cells_column_labels(everything())) %>%
  tab_style(style = cell_text(size = px(12.5), color = "grey30", 
        font = google_font("Source Sans Pro"), transform = "uppercase"), 
        locations = cells_column_spanners()) %>%
  tab_style(style = cell_text(size = px(13), color = "grey20", 
        font = google_font("Source Sans Pro"), style="italic"), 
        locations = cells_footnotes()) %>%
  tab_style(style = cell_text(size = px(13), color = "grey20", 
        font = google_font("Source Sans Pro")), 
        locations = cells_source_notes()) %>%
  tab_style(style = cell_text(color = "#203B46", 
        font = google_font("Source Sans Pro"), weight=600, transform = "uppercase"), 
        locations = cells_row_groups()) 

WESTERN U.S. REGISTERED NURSES 2020
State	Salary/Wage			Total Employed			Location
	Annual ($)	Hourly ($)	SE (%)	RN1	Healthcare2	Yearly3	Quotient
Pacific division
California	120,560	57.96	1	307,060	844,740	16,430,660	0.87
Hawaii	104,830	50.4	1.9	11,260	31,410	574,010	0.91
Oregon	96,230	46.27	1	36,840	100,230	1,806,950	0.95
Alaska	95,270	45.81	1.4	6,240	17,730	296,300	0.98
Washington	91,310	43.9	0.9	59,300	171,850	3,195,200	0.86
Mountain division
Nevada	89,750	43.15	1.9	23,420	67,590	1,250,860	0.87
Arizona	80,380	38.64	0.9	55,520	171,010	2,835,110	0.91
Colorado	77,860	37.43	0.7	52,330	144,490	2,578,000	0.95
New Mexico	75,700	36.4	1.4	17,100	47,340	785,720	1.01
Wyoming	72,600	34.9	1.1	5,010	14,010	261,690	0.89
Idaho	71,640	34.44	0.9	12,800	40,750	718,820	0.83
Montana	70,530	33.91	1.2	9,980	29,770	455,450	1.02
Utah	70,370	33.83	0.9	23,690	72,300	1,489,020	0.74
Data source: Data.World
1 Total employed registered nurses 2 Total employed healthcare, state aggregate 3 Yearly total employed, state aggregte

Northeast region table

# gt_plt_bar_stack reference: https://github.com/BjnNowak/TidyTuesday/blob/main/SC_Nurse.R
bar_hourly = df20 %>% filter(st %in% .northeast_region) %>% 
  group_by(state) %>%
  mutate(
    X25=round(hourly_25th_percentile),
    X50=round(hourly_wage_median),
    X75=round(hourly_75th_percentile)
    ) %>%
  arrange(-X75)%>%
  summarise(Hourly = list(c(X25,X50,X75)))

bar_annual = df20 %>% filter(st %in% .northeast_region) %>% 
  group_by(state) %>%
  mutate(
    X25=round(annual_25th_percentile),
    X50=round(annual_salary_median),
    X75=round(annual_75th_percentile)
  )%>%
  arrange(-X75)%>%
  summarise(Annual = list(c(X25,X50,X75)))

df20 %>% filter(st %in% .northeast_region) %>% 
  left_join(bar_hourly) %>%
  left_join(bar_annual) %>%
  arrange(desc(annual_salary_avg)) %>%
  select(state, total_employed_rn, annual_salary_avg, Annual, hourly_wage_avg, Hourly) %>%
  gt() %>%
  gt_theme_538() %>%
  #gt_merge_stack(col1 = state, col2 = st) %>%
  gt_plt_dot(annual_salary_avg,state,palette = c("#6c757d", "#dee2e6")) %>%
  gt_plt_bar_stack(
    column=Annual, palette = c("#a4243b","#05668d","#d8973c"),
    position = 'stack', labels = c("1st quartile", "Median", "3rd quartile"),
    width = 60,trim=TRUE
  ) %>%
  gt_plt_bar_stack(
    column=Hourly, palette = c("#5f0f40","#e36414","#0f4c5c"),
    position = 'stack', labels = c("1st quartile", "Median", "3rd quartile"),
    width = 60,trim=TRUE
  ) %>%
  cols_label(
    total_employed_rn = html("Employment"),
    annual_salary_avg = html("Avg"),
    hourly_wage_avg = html("Avg")
    ) %>%
  tab_spanner(
    label = "Annual Salary ($)",
    columns = annual_salary_avg:Annual
  ) %>%
  tab_spanner(
    label = "Hourly Wage ($)",
    columns = hourly_wage_avg:Hourly
  ) %>%
  tab_style(
    style = list(cell_text(align = "center")),
    locations = cells_column_labels(columns=c(annual_salary_avg,hourly_wage_avg))
  ) %>%
  tab_style(
    style = list(cell_text(align = "left")),
    locations = cells_column_labels(columns=c(Annual,Hourly))
  ) %>%
  tab_style(
    style = list(
      cell_text(font=google_font(
        name = "Libre Franklin"), weight='800',align = "left",color='black')),
    locations = cells_title(groups = "title")
  ) %>%
  tab_options(data_row.padding = px(6),
              table.font.size = "14px",
              column_labels.font.size = "10.5px",
              heading.padding = px(5),
              column_labels.padding = px(8),
              source_notes.padding = px(14),
              ) %>%
  tab_header(title = "2020 U.S. Registered Nurses in the Northeast Region", subtitle="New York has the highest number of employed registered nurses, and Massachusetts has the highest average annual salary and hourly wage in 2020.") %>%
  tab_source_note(source_note="Data source: Data.World") 

2020 U.S. Registered Nurses in the Northeast Region
New York has the highest number of employed registered nurses, and Massachusetts has the highest average annual salary and hourly wage in 2020.
state	Employment	Annual Salary ($)		Hourly Wage ($)
		Avg	1st quartile||Median||3rd quartile	Avg	1st quartile||Median||3rd quartile
Massachusetts	84030	96250		46.27
New York	178550	89760		43.16
New Jersey	78590	85720		41.21
Connecticut	33400	84850		40.79
Rhode Island	12150	82790		39.81
New Hampshire	13840	75970		36.52
Pennsylvania	146640	74170		35.66
Vermont	6810	72140		34.68
Maine	14160	71040		34.16
Data source: Data.World

Western region: annual salary

p3 = df20 %>% filter(st %in% .west_region) %>%
  select(state, annual_salary_median,annual_10th_percentile,annual_90th_percentile,
         annual_25th_percentile,annual_75th_percentile) %>% 
  arrange(desc(annual_salary_median)) %>%
  mutate(state=fct_rev(fct_inorder(state))) %>%
  ggplot() +
  geom_segment(aes(y=state, yend=state, x=annual_10th_percentile, xend=annual_90th_percentile), 
               linetype="dotted", color="white") +
  geom_segment(aes(y=state, yend=state, x=annual_25th_percentile, xend=annual_75th_percentile), color="white", size=.8) +
  geom_point(aes(y=state, x=annual_salary_median), size=2.3, color="white") +
  geom_point(aes(y=state, x=annual_25th_percentile), size=3, color="white", shape="|") +
  geom_point(aes(y=state, x=annual_75th_percentile), size=3, color="white", shape="|") +
  geom_point(aes(y=state, x=annual_10th_percentile), size=2, color="white", shape="|") +
  geom_point(aes(y=state, x=annual_90th_percentile), size=2, color="white", shape="|") +
  scale_x_continuous("Annual Salary", labels=dollar_format(scale = .001, suffix = "K"), 
                     expand=c(0,0), limits=c(50000,175000)) +
  coord_cartesian(clip="off") +
  scale_y_discrete("",expand = expansion(mult = c(0.08, .15))) +
  theme_minimal(base_size = 10) +
  theme(panel.grid.major.y=element_blank(),
        panel.grid.minor=element_blank(),
        panel.grid=element_line(size=.2, color="#343a40"),
        plot.title.position = "plot",
        plot.title=element_markdown(color="white"),
        plot.background = element_rect(fill="#212529",color=NA),
        axis.text.y=element_text(color="white", face="bold"),
        axis.text.x=element_text(color="white", size=7),
        axis.title=element_text(color="white", size=8),
        plot.margin = unit(c(.75, 2, .5, 1), "cm"),
        plot.caption=element_text(color="white", size=5.5)
        ) +
  labs(title="Annual Salary of **Registered Nurses** in Western US (2020)", 
       caption="Data from: Data.World") +
  annotate(geom="text",,size=2.8, y=13.9, x=c(76180, 93970, 118410,147830,173370), 
           label=c("10th pctl","25th pctl","Median","75th pctl","90th pctl"), color="white")

p4 = plot_usmap(include = .west_region, color="#adb5bd", fill="#212529")

p3 | inset_element(p4, align_to = "full", clip=F, on_top=T, ignore_tag = T,
                   left =0.57, bottom=.2, right=1, top=.6)

Midwest region: median annual salary

plot_usmap(data=df20, values="annual_salary_median", include=.midwest_region, color="white") +
  scale_fill_continuous_sequential(palette="PuBuGn", breaks=c(60000, 65000, 70000,75000,79540),label=dollar_format()) +
  theme_void(base_size = 8.5) +
  theme(plot.margin = unit(c(.75, 1, .55, 1), "cm"),
        legend.title=element_text(size=7.7),
        plot.title=element_markdown(size=12,hjust=.5, margin=margin(b=10)),
        plot.caption = element_text(size=6),
        legend.position = "top") +
  coord_fixed() +
  annotate(geom="richtext",label.color=NA, size=2.6,fill=NA,
           color=c("black","black","black","black","white","black","black","white","black","black","white","white"),
           x=c(-40000, -30000, 30000, 130000, 430000, 530000,660000,900000,1180000, 1450000, 1250000,780000), 
           y=c(270000,-50000, -400000, -720000, 250000, -300000, -720000, -500000,-450000,-380000,-100000,20000), 
           label=c("North Dakota<br>$68,800","South Dakota<br>$60,000","Nebraska<br>$68,010",
                   "Kansas<br>$62,550","Minnesota<br>$79,540","Iowa<br>$61,130",
                   "Missouri<br>$64,220", "Illinois<br>$72,610", "Indiana<br>$65,000",
                   "Ohio<br>$67,580","Michigan<br>$73,040","Wisconsin<br>$73,540")) +
  labs(title="**Annual Salary of Registered Nurses in the Midwest**",
       fill="Median Annual Salary in 2020", caption="Data from: Data.World") +
  guides(fill = guide_colorbar(title.position = "top", 
                                title.hjust = .5, 
                                barwidth = unit(15, "lines"), 
                                barheight = unit(.5, "lines")))

# north east grid 
northeast_grid2 = us_state_grid2 %>% filter(code %in% .northeast_region) %>%
  mutate(col = col-8,
         col=ifelse(row==1,col-1,col))

# percent change (wage and total emp)
nurses %>% 
  filter(st %in% .northeast_region) %>% 
  select(year,state,hourly_wage_median, total_employed_rn) %>%
  group_by(state) %>%
  arrange(year, .by_group=TRUE) %>%
  mutate("Hourly wage median"=(hourly_wage_median/lag(hourly_wage_median)-1),
         "Total employed RN"=(total_employed_rn/lag(total_employed_rn)-1)) %>%
  select(year, state, "Hourly wage median", "Total employed RN") %>%
  pivot_longer("Hourly wage median":"Total employed RN") %>%
  ggplot(aes(x=year, y=value, color=(name))) +
  geom_hline(yintercept=0, linetype="dashed", color="#747474") +
  geom_line(show.legend = F) +
  scale_y_continuous(labels=scales::percent) +
  scale_x_continuous(breaks=c(2000,2010,2020)) +
  facet_geo(~state, grid=northeast_grid2) +
  scale_color_manual(values=c("#f28f3b","#588b8b")) +
  #scale_color_manual(values=c("#ff8811","#eaf8bf")) +
  theme_gdocs(base_size = 5) +
  theme(panel.grid.major=element_blank(),
        axis.line.x=element_blank(),
        strip.text=element_text(size=8, color="white"),
        axis.text=element_text(color="white"),
        axis.title=element_blank(),
        plot.background = element_rect(fill="#212529", color=NA),
        plot.margin = unit(c(.5, 2, .5, 1.5), "cm"),
        panel.spacing.x = unit(2, "lines"),
        plot.title.position="plot",
        plot.title=element_text(color="white", size=11),
        plot.subtitle=element_markdown(color="#e9ecef", size=8.5),
        ) +
  labs(title="Northeast Region Registered Nurses (1998-2020)",
       subtitle= "<span style = 'color:#f28f3b;'><b>Median hourly wage<b></span> and <span style = 'color:#588b8b;'><b>total employed<b></span>, expressed in percentage change over previous year<br>")

NA

Southern region: hourly wage

library(ggfan)

south_grid2 = us_state_grid2 %>% filter(code %in% .south_region) %>%
  mutate(col=col-3, row=row-3)

fan = nurses %>% 
  mutate(st=ifelse(state=="District of Columbia","DC",st)) %>%
  filter(st %in% .south_region) %>%
  select(state, year, hourly_wage_median, contains("hourly") & contains("percentile")) %>% 
  rename("percentile_0" = "hourly_wage_median") %>%  
  pivot_longer(contains("perc")) %>% 
  mutate(percentile = parse_number(name) / 100)

ggplot(fan, aes(x = year, y = value, quantile = percentile)) +
  geom_fan() +
  geom_line(aes(group = percentile), size = 0.2, color = "white") +
  scale_x_continuous(breaks = c(2000, 2010, 2020)) +
  scale_y_continuous(labels=dollar_format()) +
  scale_fill_stepsn(colors = c("#38a3a5", "#22577a")) +
  facet_geo(~state, grid=south_grid2) +
  theme_minimal(base_size = 6, base_family = "Arial Narrow") +
  theme(legend.position = "none",
        strip.text=element_text(size=5.5, face="bold", family="Arial Narrow"),
        panel.grid.minor=element_blank(),
        axis.title=element_blank(),
        panel.grid=element_line(size=.2),
        plot.margin = unit(c(-.5, 1, .5, 1), "cm"),
        plot.title = element_text(vjust = - 16, size=11, face="bold"),
        plot.subtitle = element_text(vjust = - 26, size=8, lineheight = 1.3),
        plot.caption=element_text(hjust=.5, size=5)
        ) +
  labs(title="Registered Nurses Hourly Wages in Southern US",
       subtitle="From 1998 to 2020\n10th, 25th, 50th(median),75th and 90th percentile",
       caption="\nData source: Data.World")