start to wrangle NASS quickstats organic survey data

organic_ag_data_nass_quickstats.R

#CSV downloaded from https://quickstats.nass.usda.gov/results/12827E57-2B20-368C-A91F-1B97FF8F7B35#E2E0BF24-9F88-35DB-8526-8E47D941FDBF
library(dplyr)
library(tidyr)
nass_data <- readr::read_csv("raw_data/E2E0BF24-9F88-35DB-8526-8E47D941FDBF.csv",na = '(D)',
                             skip = 1,
                              
                             col_names = c(
                               "program", "period", "week_ending", "geo_level", "state", "state_ansi", 
                               "ag_district", "ag_district_code", "county", "county_ansi", "zip_code", 
                               "region", "watershed_code", "watershed", "commodity", "data_item", 
                               "domain", "domain_category", "value_2008", "cv_2008", "value_2011", 
                               "cv_2011", "value_2014", "cv_2014", "value_2019", "cv_2019"), 
                             col_types = cols(
                               .default = col_character(),
                               state_ansi = col_integer(),
                               cv_2008 = col_number(),
                               cv_2011 = col_number(),
                               cv_2014 = col_number(),
                               cv_2019 = col_number(),
                               value_2008 = col_number(),
                               value_2011 = col_number(),
                               value_2014 = col_number(),
                               value_2019 = col_number()
                             )) %>% 
  select(geo_level, state, state_ansi, commodity, data_item, domain, domain_category, 
         starts_with(c('value', 'cv'))) %>% 
  filter(domain_category == "ORGANIC STATUS: (NOP USDA CERTIFIED)")
  
summary(nass_data)
lapply(nass_data %>% select(-starts_with(c('value', 'cv'))), unique)
x <- pivot_longer(nass_data, 
             cols = starts_with(c('value', 'cv')), 
             names_sep = '_',
             names_to = c('.value', 'year'),
             names_transform = list(year = as.integer),
             ) 
national <- x %>% filter(geo_level == 'NATIONAL')
state <- x %>% filter(geo_level == 'STATE')

ggplot(national) + 
  geom_point(aes(year, value)) + 
  facet_wrap(~commodity + data)

  select(year, value, cv)