01 - Data preparation

01_D_data_preparation.qmd

General overview

In this document we provide all the scripts needed to process the main data and later create models and figures.

Reading libraries, data and basic functions

library(dplyr)   # data manipulation
library(tidyr)   # data tidying
library(glue)    # string interpolation
library(here)    # constructing file paths
library(readr)   # reading CSV files

# create period of 50 years
floor_period = function(value){ return(value - value %% 50) }

Data containing Name Bearing Types (simply name bearers) flow. This data frame contains the flow of name bearers among countries. The source country of the name bearers is represented by its three letter acronym (country_type) and its respective world bank region (region_type). The housing location is represented by three letter code of the country (contry_museum) and its respective world bank region of the museum (region_museum). name bearers flowing is represented by 50-year time intervals. The total number of name bearers flowing from one region/country to another is represented by the column n. These names and meanings are consistent throughout all data sets

# Flow by each period, region and country
flow_period_region_country <- readr::read_csv(here::here("data", "raw", "flow_period_region_country.csv"))

Load data

# 1

# Flow by each period - total number of name bearers by period
flow_period <- flow_period_region_country |> 
  dplyr::group_by(period) |> 
  dplyr::count(wt = n) |> 
  dplyr::ungroup() |>
  dplyr::add_count(name = "total_period", wt = n)

# Flow by each region - total number of name bearers among regions
flow_region <- flow_period_region_country |> 
  dplyr::group_by(region_type, region_museum) |> 
  dplyr::count(wt = n) |> 
  dplyr::ungroup() |>
  dplyr::group_by(region_type) |>
  dplyr::add_count(name = "total_region_type", wt = n) |> 
  dplyr::ungroup()

readr::write_csv(flow_region, here::here("data","processed", "flow_region.csv"))

# Flow among countries
flow_country <- flow_period_region_country |> 
  dplyr::group_by(country_type, country_museum) |> 
  dplyr::count(wt = n) |> 
  dplyr::ungroup() |>
  dplyr::group_by(country_type) |>
  dplyr::add_count(name = "total_country_type", wt = n) |> 
  dplyr::ungroup()

readr::write_csv(flow_country, here::here("data","processed", "flow_country.csv"))


# 2

# Flow by each period and region - add total from period and region_type
flow_period_region <- flow_period_region_country |> 
  dplyr::group_by(period, region_type, region_museum) |> 
  dplyr::count(wt = n) |> 
  dplyr::ungroup()|>
  dplyr::group_by(period, region_type) |>
  dplyr::add_count(name = "total_period_region_type", wt = n) |> 
  dplyr::ungroup()

readr::write_csv(flow_period_region, here::here("data","processed", "flow_period_region.csv"))

# Flow by each region and country - add total from country_type
flow_region_country <- flow_period_region_country |> 
  dplyr::group_by(region_type, region_museum,
                  country_type, country_museum) |> 
  dplyr::count(wt = n) |> 
  dplyr::ungroup() |>
  dplyr::group_by(country_type) |>
  dplyr::add_count(name = "total_country_type", wt = n) |> 
  dplyr::ungroup()

# Proportions

# Flow by each period and region - add proportions
flow_period_region_prop <- flow_period_region |>
  # ungroup
  dplyr::ungroup() |>
  # group by period and museum region
  dplyr::group_by(period, region_museum) |>
  # count how many name bearerss were deposited in the museums of this region during the period
  dplyr::add_count(wt = n, name = "total_period_region_museum") |>
  # ungroup
  dplyr::ungroup() |>
  # group by period
  dplyr::group_by(period) |>
  # count how many name bearerss were described during the period
  dplyr::add_count(wt = n, name = "total_period") |>
  # filter only within each region
  dplyr::filter(region_type == region_museum) |>
  #add prop_DC (Domestic Contribution) and prop_DR (Domestic Retention)
  dplyr::mutate(prop_DC = n/total_period_region_museum,
                prop_DR = n/total_period_region_type) |> 
  dplyr::ungroup() |> 
  dplyr::mutate(prop_DC = ifelse(is.na(prop_DC), 0, prop_DC))

readr::write_csv(flow_period_region_prop, here::here("data","processed", "flow_period_region_prop.csv"))

# Flow by each period and region - add proportions
flow_region_prop <- flow_region |>
  # ungroup
  dplyr::ungroup() |>
  # group by period and museum region
  dplyr::group_by(region_museum) |>
  # count how many name bearers were deposited in the museums of this region
  dplyr::add_count(wt = n, name = "total_region_museum") |>
  # ungroup
  dplyr::ungroup() |>
  # filter only within each region
  dplyr::filter(region_type == region_museum) |>
  #add prop_DC (Domestic Contribution) and prop_DR (Domestic Retention)
  dplyr::mutate(prop_DC = n/total_region_museum,
                prop_DR = n/total_region_type)

readr::write_csv(flow_region_prop, here::here("data","processed", "flow_region_prop.csv"))

Museum data

museum_data <- readr::read_csv(here::here("data", "raw", "museum_data.csv"))

infra_museum <-
  museum_data |>
  dplyr::group_by(country_museum) |>
  dplyr::count(name = "n.museums")

readr::write_csv(infra_museum, here::here("data","processed", "infra_museum.csv"))

Native and Types species by country list

# Native
spp_native_distribution <- readr::read_csv(here::here("data", "raw", "spp_native_distribution.csv")) 

df_country_native <- spp_native_distribution |>
  dplyr::group_by(country_distribution, region_distribution) |>
  dplyr::count(name = "native.richness") |>
  dplyr::ungroup()

readr::write_csv(df_country_native, here::here("data","processed", "df_country_native.csv"))

# Types
spp_type_distribution <- readr::read_csv(here::here("data", "raw", "spp_type_distribution.csv")) 

df_country_type <- spp_type_distribution |>
  dplyr::group_by(country_museum, region_museum) |>
  dplyr::count(name = "type_richness") |>
  dplyr::ungroup()

readr::write_csv(df_country_type, here::here("data","processed", "df_country_type.csv"))

Bio-Dem

We downloaded data from Bio-Dem data base and saved them as csv files. Here I read these files that contain information on GDP per capta (e_migdppc), total number of occurrence species records from gbif (records), records per area from gbif (records_per_area), years since independence for each country (yearsSinceIndependence). Country names is represented by iso3c code - an unique three letters code that identify each country.

df_bio_dem <- readr::read_csv(file = here::here("data", "raw", "bio-dem_data.csv"))