R Notebook

This is an R Markdown— title: “APratt Pew Assignment” output: html_document —

R Markdown

This is an R Markdown document.

library(tidyverse)
pew <- read_csv("January 3-10, 2018 - Core Trends Survey/January 3-10, 2018 - Core Trends Survey - CSV.csv")

Parsed with column specification:
cols(
  .default = col_double(),
  usr = col_character(),
  `pial11ao@` = col_character()
)
See spec(...) for full column specifications.

pew <- pew %>% 
  mutate(web1e = as.factor(web1e)) %>% 
  mutate(youtube_use = fct_recode(web1e, "Yes" = "1", "No" = "2", NULL = "8", NULL = "9")) 

pew %>% 
  drop_na(youtube_use) %>% 
  count(youtube_use)

Perhaps unsurprisingly, more people answered “Yes” to using YouTube than “No.” Cat vids, man.

pew <- pew %>% 
  mutate(educ2 = as.factor(educ2)) %>% 
  mutate(education_level = fct_recode(educ2, 
                                      "Less than HS" = "1", 
                                      "Some HS" = "2", 
                                      "HS graduate" = "3", 
                                      "Some college" = "4", 
                                      "Associate degree" = "5", 
                                      "College degree" = "6", 
                                      "Some grad school" = "7", 
                                      "Grad degree" = "8", 
                                      NULL = "98", 
                                      NULL = "99"))

pew %>% 
  drop_na(education_level) %>% 
  count(education_level)

This table shows that the highest number of YouTube users have a college degree and are therefore watching cat videos at work. Losers.

pew %>% 
  drop_na(education_level, youtube_use) %>% 
  count(education_level, youtube_use)

Here’s the two factors run simultaneously, showing that college grads really do watch more cat videos than the rest of us.

pew %>% 
  drop_na(youtube_use) %>% 
  ggplot(aes(x = youtube_use, fill = sex)) +
  geom_bar()+ 
  scale_fill_viridis_d() +
  coord_flip()

Shall we start with a basic yes/no graph? R Studio has a personal vendetta against me and will not publish my graphs in color. Just imagine it pretty, okay? Okay.

pew %>% 
  drop_na(youtube_use) %>%
  drop_na(education_level) %>% 
  ggplot(aes(x = education_level, fill = youtube_use)) +
  geom_bar(position = "fill") +
  scale_fill_viridis_d() +
  coord_flip() +
  theme_minimal() +
  labs(x = "Level of education", y = "Percentage", fill = "Do you use YouTube?", title = "YouTube Usage by Education Level")

Okay, this graph showed up in color so whatever. Look at this one instead.

pew <- pew %>% 
  mutate(education_level_simple = fct_collapse(education_level, 
                                               "no_degree" = c("HS graduate", 
                                                                        "Some HS", 
                                                                        "Less than HS",
                                                               "Some college"),
                                               "degree" = 
                                                 c("Associate degree", 
                                                   "College degree", 
                                                   "Some grad school", 
                                                   "Grad degree")))

pew %>% 
  drop_na(education_level_simple, youtube_use) %>% 
  count(education_level_simple, youtube_use)

People apparently love YouTube. 1434 “yes” responses versus 484 “no.”

pew %>% 
  drop_na(youtube_use) %>%
  drop_na(education_level_simple) %>% 
  ggplot(aes(x = education_level_simple, fill = youtube_use)) +
  geom_bar(position = "fill") +
  scale_fill_viridis_d() +
  coord_flip() +
  theme_minimal() +
  labs(x = "Level of education", y = "Percentage", fill = "Do you use YouTube?", title = "YouTube Usage by Education Level")

pew <- pew %>% 
  mutate(emplnw = as.factor(emplnw)) %>% 
  mutate(employement = fct_recode(emplnw, 
                                  "Employed full time" = "1", 
                                  "Employed part-time" = "2",
                                  "Retired" = "3",
                                  "Not employed" = "4",
                                  "Self-employed" = "5", 
                                  "Disabled" = "6", 
                                  "Student" = "7", 
                                  NULL = "8", 
                                  NULL = "98", 
                                  NULL = "99"))

pew %>% 
  drop_na(emplnw) %>% 
  count(employement)

Factor `employement` contains implicit NA, consider using `forcats::fct_explicit_na`

Since I made fun of employed people watching the most YouTube, let’s back it up. First up is a factor and recode. Now let’s cross it over with actual use.

pew %>% 
  drop_na(employement, youtube_use) %>% 
  count(employement, youtube_use)

What are they watching? Why?

pew <- pew %>% 
  mutate(employement_simple = fct_collapse(employement, 
                                               "employed" = c("Employed full time", 
                                                                        "Employed part-time"),
                                               "not employed" = 
                                                 c("Retired", 
                                                   "Not employed", 
                                                   "Self-employed",
                                                   "Student",
                                                   "Disabled")))
pew %>% 
  drop_na(employement_simple, youtube_use) %>% 
  count(employement_simple, youtube_use)

Almost twice as many employed people. Let’s see a graph and then I’ll leave it alone.

pew %>% 
  drop_na(youtube_use) %>%
  drop_na(employement_simple) %>% 
  ggplot(aes(x = employement_simple, fill = youtube_use)) +
  geom_bar(position = "fill") +
  scale_fill_viridis_d() +
  coord_flip() +
  theme_minimal() +
  labs(x = "Employment Status", y = "Percentage", fill = "Do you use YouTube?", title = "YouTube Usage by Employment Status")

This is so great, I love it.