02_traits.Rmd

name: traits
# Traits

```{r, echo = F}
df <- readxl::read_xlsx("raw_data/Looks vs Personality.xlsx") %>%
  mutate(country = str_replace(Nationality, "n$", "")) %>%
  mutate(country = recode(country,
    America = "USA",
    Arabia = "Saudi Arabia",
    British = "United Kingdom",
    Danish = "Denmark",
    Egyptia = "Egypt",
    Filipino = "Philippines",
    Finnish = "Finland",
    French = "France",
    Germa = "Germany",
    Norwegia = "Norway",
    Singaporea = "Singapore",
    Swedish = "Sweden",
    Thai = "Thailand",
    Vietnamese = "Vietnam"
  )) %>%
  mutate(country_name = 
           countrycode::countrycode(country, 
                                    "country.name", 
                                    "country.name")) %>%
  mutate(Percent = Percentage * 100) %>%
  mutate(
    Question_short =
      recode(Question,
        `They have a personality I like` = "Personality",
        `They have a sense of humour I like` = "Sense of humor",
        `They are intelligent` = "Intellegence",
        `They have similar interests to me` = "Has similar interests",
        `They are good looking` = "Looks",
        `They have/make a decent amount of money` = "Money"
      )
  ) %>%
  mutate(`Rank (text)`)


tmp <- df %>%
  group_by(Question, Question_short) %>%
  summarise(max = mean(Percentage * (7 - `Rank (number)`))) %>%
  arrange(-max)

df <- df %>%
  mutate(Question = factor(Question, tmp$Question)) %>%
  mutate(Question_str_wrap = factor(
    str_wrap(Question, 12),
    str_wrap(tmp$Question, 12)
  )) %>%
  mutate(Question_short_wrap = factor(
    str_wrap(Question_short, 12),
    str_wrap(tmp$Question_short, 12)
  ))

world <- df %>%
  group_by(Gender, Question_short, Question_short_wrap, `Rank (text)`, `Rank (number)`) %>%
  summarise(n = sum(Unweighted_Sample * Percentage)) %>%
  group_by(Gender, Question_short, Question_short_wrap) %>%
  mutate(Percent = 100 * n / sum(n)) %>% 
  mutate(`Rank (text)` = factor(`Rank (text)`, levels = c("Ranked sixth", "Ranked fifth", "Ranked fourth", "Ranked third", "Ranked second", "Ranked first")))


world_surved <- df %>%
  group_by(Gender, Nationality) %>%
  summarise(n = mean(Unweighted_Sample)) %>%
  group_by(Gender) %>%
  summarise(n = sum(n))

```

A random sample from the data set: 
```{r, echo = F}
knitr::kable(sample_n(world %>% ungroup() %>% select(-Question_short_wrap), 5), format = "html")
```

---

```{r traits, fig.height=6, fig.width=12, eval=F, echo=F}
ggplot(data = world) + 
  aes(x = Question_short_wrap) +
  aes(y = Percent) +
  aes(fill = `Rank (text)`) +
  facet_grid(Gender ~ .) +
  geom_col() +
  coord_flip() +
  scale_fill_manual(
    values = colorRampPalette(
      RColorBrewer::brewer.pal(n = 9, name = "Purples"))(6), 
                    guide = guide_legend(reverse = TRUE)) +
  labs(fill = "") +
  xlab("") +
  labs(title = "Why do I love thee?  Let me rank the traits... \nHow 10,689 men and 11,370 women across 20 countries rank romantic partner trait importance") +
  labs(subtitle = "Data Source: @mattsmithetc and @YouGov | Design: Gina Reynolds") 
```

```{r, echo = F, warning=F, message=F, eval = T, fig.show='hide'}
get_what_save_what <- "traits"
eval(parse(text = paste(knitr:::knit_code$get(get_what_save_what), collapse = "")))
ggsave(paste0("figures/", get_what_save_what, ".png"), dpi = 300)
```

`r apply_reveal("traits")`

---