manipulating_data.R

library(tidyverse)
surveys <- read_csv("data_raw/portal_data_joined.csv")
str(surveys)

select(surveys, plot_id, species_id, weight)
select(surveys, -record_id, -species_id)

filter(surveys, year == 1995)

surveys2 <- filter(surveys, weight < 5)
surveys_sml <- select(surveys2, species_id, sex, weight)
surveys_sml <- select(filter(surveys, weight < 5), 
                      species_id, sex, weight)

# to get a pipe do Ctrl - Shift - M
# or on a Mac Cmd - Shift - M
# or type %>%
surveys %>%
  filter(weight < 5) %>%
  select(species_id, sex, weight)

surveys %>% 
  select(plot_id, sex)

select(surveys, plot_id, sex)

surveys %>%
  filter(weight < 5) %>%
  select(species_id, sex, weight)

surveys_sml <- surveys %>%
                filter(weight < 5) %>%
                select(species_id, sex, weight)

surveys %>% 
  filter(year < 1995) %>% 
  select(year, sex, weight)

surveys %>% 
  mutate(weight_kg = weight / 1000)

str(surveys)

new_tbl <- surveys %>% 
  mutate(weight_kg = weight / 1000)

View(new_tbl)

# this makes an error because == is a comparison
new_tbl <- surveys %>% 
  mutate(weight_kg == weight / 1000)

surveys %>% 
  mutate(weight_kg = weight / 1000,
         weight_lb = weight_kg * 2.2) %>% 
  head()

surveys %>%
  filter(!is.na(weight)) %>% 
  mutate(weight_kg = weight / 1000,
         weight_lb = weight_kg * 2.2) %>% 
  head()

# form1
surveys %>% 
  filter(!year < 1995) %>% 
  head()

# form2 - equivalent to form1
surveys %>% 
  filter(year >= 1995) %>% 
  head()


surveys %>% 
  group_by(sex) %>% 
  summarize(mean_weight = mean(weight, na.rm = TRUE))

surveys %>% 
  group_by(sex)
  

surveys %>% 
  summarize(mean_weight = mean(weight))

surveys %>% 
  summarize(mean_weight = mean(weight, na.rm = TRUE))

surveys %>% 
  group_by(sex) %>% 
  summarize(mean_weight = mean(weight, na.rm = TRUE))

surveys %>% 
  group_by(sex, species_id) %>% 
  summarize(mean_weight = mean(weight, na.rm = TRUE))

data_overview <- surveys %>% 
  group_by(sex, species_id) %>% 
  summarize(mean_weight = mean(weight, na.rm = TRUE))

View(data_overview)


surveys %>% 
  filter(!is.na(weight)) %>% 
  group_by(sex, species_id) %>% 
  summarize(mean_weight = mean(weight)) %>% 
  print(n = 64)

surveys %>% 
  filter(!is.na(weight)) %>% 
  group_by(sex, species_id) %>% 
  summarize(mean_weight = mean(weight),
            min_weight = min(weight)) %>% 
  arrange(min_weight)

surveys %>% 
  filter(!is.na(weight)) %>% 
  group_by(sex, species_id) %>% 
  summarize(mean_weight = mean(weight),
            min_weight = min(weight)) %>% 
  arrange(min_weight, mean_weight)

surveys %>% 
  filter(!is.na(weight)) %>% 
  group_by(sex, species_id) %>% 
  summarize(mean_weight = mean(weight),
            min_weight = min(weight)) %>% 
  arrange(desc(min_weight))

# Counting data by criteria
surveys %>% 
  count(sex)

surveys %>% 
  count(year < 1995)

surveys %>% 
  count(sex, species_id)

surveys %>% 
  count(sex, species_id) %>% 
  arrange(species_id, desc(n))