Module 11: Apply it to your data 10

Import your data

data <- read_excel("CliffordSalmonidDataApply10.xlsx")

Chapter 15

Create a factor

data <- data %>%
  mutate(region = factor(region))

data

## # A tibble: 2,808 × 9
##    species date                geo_group region losses   dead discarded escaped
##    <chr>   <dttm>              <chr>     <fct>   <dbl>  <dbl>     <dbl>   <dbl>
##  1 salmon  2020-01-01 00:00:00 area      1       31425  28126      3299       0
##  2 salmon  2020-01-01 00:00:00 area      2      324116 277888     46113       0
##  3 salmon  2020-01-01 00:00:00 area      3      844829 776983     63770       0
##  4 salmon  2020-01-01 00:00:00 area      4      676852 623159     51823       0
##  5 salmon  2020-01-01 00:00:00 area      5      109269  97627     11424       0
##  6 salmon  2020-01-01 00:00:00 area      6      548921 531193     15710       0
##  7 salmon  2020-01-01 00:00:00 area      7      231487 228847      2640       0
##  8 salmon  2020-01-01 00:00:00 area      8      442659 431218     11228       0
##  9 salmon  2020-01-01 00:00:00 area      9      311127 308298      2212       0
## 10 salmon  2020-01-01 00:00:00 area      10     288849 281808      3120       0
## # ℹ 2,798 more rows
## # ℹ 1 more variable: other <dbl>

Modify factor order

Make two bar charts here - one before ordering another after

data %>%
  group_by(region) %>%
  summarise(total_losses = sum(losses, na.rm = TRUE)) %>%
  ggplot(aes(x = total_losses, y = region)) +
  geom_col() +
  labs(title = "Total Losses by Region (Unordered)",
       x = "Total Losses", 
       y = "Region")

data %>%
  group_by(region) %>%
  summarise(total_losses = sum(losses, na.rm = TRUE)) %>%
  ggplot(aes(x = total_losses, 
             y = fct_reorder(.f = region, .x = total_losses))) +
  geom_col() +
  labs(title = "Total Losses by Region (Ordered)",
       x = "Total Losses", 
       y = NULL)

Modify factor levels

Show examples of three functions:

• fct_recode

data %>%
  mutate(region_rec = fct_recode(region, 
                                 "Region One" = "1", 
                                 "Region Two" = "2")) %>%
  select(region, region_rec) %>%
  filter(region %in% c("1", "2"))

## # A tibble: 288 × 2
##    region region_rec
##    <fct>  <fct>     
##  1 1      Region One
##  2 2      Region Two
##  3 1      Region One
##  4 2      Region Two
##  5 1      Region One
##  6 2      Region Two
##  7 1      Region One
##  8 2      Region Two
##  9 1      Region One
## 10 2      Region Two
## # ℹ 278 more rows

• fct_collapse

data %>%
  mutate(region_col = fct_collapse(region, 
                                   "Zone A" = c("1", "2", "3"),
                                   "Zone B" = c("4", "5", "6"))) %>%
  select(region, region_col)

## # A tibble: 2,808 × 2
##    region region_col
##    <fct>  <fct>     
##  1 1      Zone A    
##  2 2      Zone A    
##  3 3      Zone A    
##  4 4      Zone B    
##  5 5      Zone B    
##  6 6      Zone B    
##  7 7      7         
##  8 8      8         
##  9 9      9         
## 10 10     10        
## # ℹ 2,798 more rows

• fct_lump

data %>% 
  count(region)

## # A tibble: 23 × 2
##    region     n
##    <fct>  <int>
##  1 1        144
##  2 10       144
##  3 11        72
##  4 12        72
##  5 13        72
##  6 2        144
##  7 3        144
##  8 4        144
##  9 5        144
## 10 6        144
## # ℹ 13 more rows

data %>% 
  mutate(region_lump = fct_lump(region, n = 5)) %>% 
  distinct(region_lump)

## # A tibble: 17 × 1
##    region_lump    
##    <fct>          
##  1 1              
##  2 2              
##  3 3              
##  4 4              
##  5 5              
##  6 6              
##  7 Other          
##  8 9              
##  9 10             
## 10 Agder          
## 11 Møre og Romsdal
## 12 Nordland       
## 13 Norge          
## 14 Rogaland       
## 15 Troms          
## 16 Trøndelag      
## 17 Vestland

Chapter 16

No need to do anything here.