Chapter 15 Factors

Creating factors

General Social Survey

gss_cat
## # A tibble: 21,483 × 9
##     year marital         age race  rincome        partyid    relig denom tvhours
##    <int> <fct>         <int> <fct> <fct>          <fct>      <fct> <fct>   <int>
##  1  2000 Never married    26 White $8000 to 9999  Ind,near … Prot… Sout…      12
##  2  2000 Divorced         48 White $8000 to 9999  Not str r… Prot… Bapt…      NA
##  3  2000 Widowed          67 White Not applicable Independe… Prot… No d…       2
##  4  2000 Never married    39 White Not applicable Ind,near … Orth… Not …       4
##  5  2000 Divorced         25 White Not applicable Not str d… None  Not …       1
##  6  2000 Married          25 White $20000 - 24999 Strong de… Prot… Sout…      NA
##  7  2000 Never married    36 White $25000 or more Not str r… Chri… Not …       3
##  8  2000 Divorced         44 White $7000 to 7999  Ind,near … Prot… Luth…      NA
##  9  2000 Married          44 White $25000 or more Not str d… Prot… Other       0
## 10  2000 Married          47 White $25000 or more Strong re… Prot… Sout…       3
## # ℹ 21,473 more rows

Modifying factor order

Unordered factor levels

# Transform data: calculate avergae tv hours by religion
tvhours_by_relig <- gss_cat %>%
    
    group_by(relig) %>%
    summarise(avg_tvhours = mean(tvhours, na.rm = TRUE))

tvhours_by_relig
## # A tibble: 15 × 2
##    relig                   avg_tvhours
##    <fct>                         <dbl>
##  1 No answer                      2.72
##  2 Don't know                     4.62
##  3 Inter-nondenominational        2.87
##  4 Native american                3.46
##  5 Christian                      2.79
##  6 Orthodox-christian             2.42
##  7 Moslem/islam                   2.44
##  8 Other eastern                  1.67
##  9 Hinduism                       1.89
## 10 Buddhism                       2.38
## 11 Other                          2.73
## 12 None                           2.71
## 13 Jewish                         2.52
## 14 Catholic                       2.96
## 15 Protestant                     3.15
# Plot
tvhours_by_relig %>%
    
    ggplot(aes(x = avg_tvhours, y = relig)) +
    geom_point()

Ordered factor levels

tvhours_by_relig %>%
    
    ggplot(aes(x = avg_tvhours, y = fct_reorder(.f = relig, .x = avg_tvhours))) +
    geom_point() +

# Labeling
    labs(y = NULL, x = "Mean Daily Hours Watching TV")

Moving a single level to the front

tvhours_by_relig %>%
    
    ggplot(aes(x = avg_tvhours, 
               y = fct_reorder(.f = relig, .x = avg_tvhours) %>% fct_relevel("Don't know")))+
    geom_point() +

# Labeling
    labs(y = NULL, x = "Mean Daily Hours Watching TV")

Modifying factor levels

gss_cat %>% distinct(race)
## # A tibble: 3 × 1
##   race 
##   <fct>
## 1 White
## 2 Black
## 3 Other
# Recode
gss_cat %>%
    
    # Rename levels
    mutate(race_rev = fct_recode(race, "African American" = "Black")) %>%
    select(race, race_rev) %>%
    filter(race == "Black")
## # A tibble: 3,129 × 2
##    race  race_rev        
##    <fct> <fct>           
##  1 Black African American
##  2 Black African American
##  3 Black African American
##  4 Black African American
##  5 Black African American
##  6 Black African American
##  7 Black African American
##  8 Black African American
##  9 Black African American
## 10 Black African American
## # ℹ 3,119 more rows
# Collapse multiple levels into one
gss_cat %>%
    
    mutate(race_col = fct_collapse(race, "Minority" = c("Black", "Other"))) %>%
    select(race, race_col) %>%
    filter(race != "White")
## # A tibble: 5,088 × 2
##    race  race_col
##    <fct> <fct>   
##  1 Black Minority
##  2 Black Minority
##  3 Black Minority
##  4 Other Minority
##  5 Black Minority
##  6 Other Minority
##  7 Black Minority
##  8 Other Minority
##  9 Black Minority
## 10 Black Minority
## # ℹ 5,078 more rows
# Lump small levels into other levels
gss_cat %>% count(race)
## # A tibble: 3 × 2
##   race      n
##   <fct> <int>
## 1 Other  1959
## 2 Black  3129
## 3 White 16395
gss_cat %>% mutate(race_lump = fct_lump(race)) %>% distinct(race_lump)
## # A tibble: 2 × 1
##   race_lump
##   <fct>    
## 1 White    
## 2 Other

Chapter 16 Dates and Times

Creating dates/times

# From Strings
"2024/04/09" %>% ymd()
## [1] "2024-04-09"
# From numbers
20240409 %>% ymd()
## [1] "2024-04-09"
"2024-04-09 4-44-31" %>% ymd_hms()
## [1] "2024-04-09 04:44:31 UTC"
Individual Components
flights %>%
    select(year:day, hour, minute) %>%
    mutate(departure = make_datetime(year = year, month = month, day = day, hour = hour, min = minute))
## # A tibble: 336,776 × 6
##     year month   day  hour minute departure          
##    <int> <int> <int> <dbl>  <dbl> <dttm>             
##  1  2013     1     1     5     15 2013-01-01 05:15:00
##  2  2013     1     1     5     29 2013-01-01 05:29:00
##  3  2013     1     1     5     40 2013-01-01 05:40:00
##  4  2013     1     1     5     45 2013-01-01 05:45:00
##  5  2013     1     1     6      0 2013-01-01 06:00:00
##  6  2013     1     1     5     58 2013-01-01 05:58:00
##  7  2013     1     1     6      0 2013-01-01 06:00:00
##  8  2013     1     1     6      0 2013-01-01 06:00:00
##  9  2013     1     1     6      0 2013-01-01 06:00:00
## 10  2013     1     1     6      0 2013-01-01 06:00:00
## # ℹ 336,766 more rows

From other types

today() %>% as_datetime()
## [1] "2024-04-09 UTC"
# From date-time to date
now() %>% as_date()
## [1] "2024-04-09"

Date-time components

Getting components

date_time <- ymd_hms("2024-04-09 18-18-18")
date_time
## [1] "2024-04-09 18:18:18 UTC"
year(date_time)
## [1] 2024
month(date_time, label = TRUE)
## [1] Apr
## 12 Levels: Jan < Feb < Mar < Apr < May < Jun < Jul < Aug < Sep < ... < Dec
yday(date_time)
## [1] 100
mday(date_time)
## [1] 9
wday(date_time, label = TRUE, abbr = FALSE)
## [1] Tuesday
## 7 Levels: Sunday < Monday < Tuesday < Wednesday < Thursday < ... < Saturday
make_datetime_100 <- function(year, month, day, time) {
  make_datetime(year, month, day, time %/% 100, time %% 100)
}

flights_dt <- flights %>% 
  filter(!is.na(dep_time), !is.na(arr_time)) %>% 
  mutate(
    dep_time = make_datetime_100(year, month, day, dep_time),
    arr_time = make_datetime_100(year, month, day, arr_time),
    sched_dep_time = make_datetime_100(year, month, day, sched_dep_time),
    sched_arr_time = make_datetime_100(year, month, day, sched_arr_time)
  ) %>% 
  select(origin, dest, ends_with("delay"), ends_with("time"))
flights_dt %>%
    
    transmute(wday = wday(dep_time, label = TRUE)) %>%
    
    ggplot(aes(wday)) +
    geom_bar()

Rounding

# floor_date for rounding down
flights_dt %>%
    
    mutate(week = floor_date(dep_time, "month")) %>%
    select(dep_time, week) %>%
    sample_n(10)
## # A tibble: 10 × 2
##    dep_time            week               
##    <dttm>              <dttm>             
##  1 2013-12-31 05:55:00 2013-12-01 00:00:00
##  2 2013-04-04 13:29:00 2013-04-01 00:00:00
##  3 2013-11-01 09:51:00 2013-11-01 00:00:00
##  4 2013-04-15 15:37:00 2013-04-01 00:00:00
##  5 2013-06-13 07:57:00 2013-06-01 00:00:00
##  6 2013-03-19 16:25:00 2013-03-01 00:00:00
##  7 2013-05-13 15:58:00 2013-05-01 00:00:00
##  8 2013-04-30 09:52:00 2013-04-01 00:00:00
##  9 2013-10-22 18:40:00 2013-10-01 00:00:00
## 10 2013-11-20 10:53:00 2013-11-01 00:00:00
# ceiling_date for rounding up
flights_dt %>%
    
    mutate(week = ceiling_date(dep_time, "month")) %>%
    select(dep_time, week) %>%
    sample_n(10)
## # A tibble: 10 × 2
##    dep_time            week               
##    <dttm>              <dttm>             
##  1 2013-10-21 17:35:00 2013-11-01 00:00:00
##  2 2013-11-01 15:18:00 2013-12-01 00:00:00
##  3 2013-05-24 09:36:00 2013-06-01 00:00:00
##  4 2013-01-06 09:00:00 2013-02-01 00:00:00
##  5 2013-05-25 09:32:00 2013-06-01 00:00:00
##  6 2013-09-04 13:27:00 2013-10-01 00:00:00
##  7 2013-07-15 17:11:00 2013-08-01 00:00:00
##  8 2013-09-18 14:51:00 2013-10-01 00:00:00
##  9 2013-02-11 17:04:00 2013-03-01 00:00:00
## 10 2013-08-17 19:45:00 2013-09-01 00:00:00

Setting components

flights_dt %>%
    mutate(dep_hour = update(dep_time, yday = 1)) %>%
    select(dep_time, dep_hour) %>%
    sample_n(10)
## # A tibble: 10 × 2
##    dep_time            dep_hour           
##    <dttm>              <dttm>             
##  1 2013-11-11 14:45:00 2013-01-01 14:45:00
##  2 2013-08-02 12:52:00 2013-01-01 12:52:00
##  3 2013-07-05 06:00:00 2013-01-01 06:00:00
##  4 2013-08-13 18:25:00 2013-01-01 18:25:00
##  5 2013-05-10 06:35:00 2013-01-01 06:35:00
##  6 2013-03-19 15:55:00 2013-01-01 15:55:00
##  7 2013-12-06 19:38:00 2013-01-01 19:38:00
##  8 2013-08-28 21:46:00 2013-01-01 21:46:00
##  9 2013-07-01 00:29:00 2013-01-01 00:29:00
## 10 2013-05-27 21:25:00 2013-01-01 21:25:00

Time spans