7 Descriptive statistics

Cette section vise à effectuer l’analyse descriptive de l’étude.

7.1 Statistiques descriptive des aires protégées

Code

library(tidyverse) # Manipulation et visualisation des données
library(writexl) # Pour faire une sortie sous Excel
library(psych) # Pour des analyses statistiques descriptives et psychométrique 
library(knitr) # génération de rapports dynamiques 
library(sf) # Analyse spatiale
library(readr)

# On charge les données
wdpa_before_2008 <- st_read("data/derived/wdpa_before_2008.csv")

Reading layer `wdpa_before_2008' from data source 
  `C:\Users\irian\Documents\Analyse de données\PA-livelihood-impact-dhs\data\derived\wdpa_before_2008.csv' 
  using driver `CSV'

Code

wdpa_from_2008 <- read.csv("data/derived/wdpa_from_2008.csv")

# Fonction utilitaire pour produire les statistiques d'une période
get_descriptive_stats <- function(data, periode_label) {
  rep_area_num <- as.numeric(data$REP_AREA)
  
  stats <- psych::describe(rep_area_num)[, c("n", "mean", "sd", "median", "min", "max")]
  
  min_name <- data %>%
    filter(rep_area_num == min(rep_area_num, na.rm = TRUE)) %>%
    pull(ORIG_NAME) %>%
    unique() %>%
    paste(collapse = "; ")
  
  max_name <- data %>%
    filter(rep_area_num == max(rep_area_num, na.rm = TRUE)) %>%
    pull(ORIG_NAME) %>%
    unique() %>%
    paste(collapse = "; ")
  
  stats %>%
    as_tibble() %>%
    mutate(
      periode = periode_label,
      aire_min = min_name,
      aire_max = max_name
    )
}

# Appliquer la fonction aux deux périodes
desc_AP_before <- get_descriptive_stats(wdpa_before_2008, "Before 2008")
desc_AP_from   <- get_descriptive_stats(wdpa_from_2008, "From 2008")


# Fusion des deux tableaux
desc_AP_combined <- bind_rows(desc_AP_before, desc_AP_from) %>%
  arrange(mean) %>%
  select(periode, n, mean, sd, median, min, aire_min, max, aire_max)

# Affichage du tableau 
kable(desc_AP_combined, digits = 2, caption = "Statistiques descriptives des aires protégées (REP_AREA)")

Statistiques descriptives des aires protégées (REP_AREA)
periode	n	mean	sd	median	min	aire_min	max	aire_max
From 2008	103	661.65	1100.97	222.88	0.28	Nosy Antsoha	5688.62	Complexe des AP Ambohimirahavavy Marivorahona
Before 2008	34	3068.49	13336.03	253.93	0.05	Parc de Tsarasaotra	78139.00	Ambatovaky

Code

# Enregistrement du tableau
write_csv(desc_AP_combined, "data/derived/tableau_descriptive des aires protégée.csv")

# Fonction pour produire les 10 plus petites aires protégées
get_top_small_aires <- function(df, periode, top_n = 10) {
  df %>%
    mutate(REP_AREA = as.numeric(REP_AREA)) %>%
    arrange(REP_AREA) %>%
    slice_head(n = top_n) %>%
    mutate(
      rank_number = row_number(),
      periode = periode
    ) %>% 
    select(rank_number, ORIG_NAME, REP_AREA, periode) 
}

# Appliquer aux deux périodes
top10_before_2008 <- get_top_small_aires(wdpa_before_2008, "Before 2008")
top10_from_2008  <- get_top_small_aires(wdpa_from_2008, "From 2008")

# Fusionner les tableaux
top10_combined <- bind_rows(top10_before_2008, top10_from_2008)

# Affichage
kable(top10_combined, digits = 2,
      caption = "Statistiques descriptives des aires protégées (REP_AREA) avec top 10 plus petites")

Statistiques descriptives des aires protégées (REP_AREA) avec top 10 plus petites
rank_number	ORIG_NAME	REP_AREA	periode
1	Parc de Tsarasaotra	0.05	Before 2008
2	Betampona	22.28	Before 2008
3	Ivohibe	34.53	Before 2008
4	Beza Mahafaly	42.00	Before 2008
5	Bora	48.41	Before 2008
6	Manombo	53.20	Before 2008
7	Ambohitantely	56.00	Before 2008
8	Maningoza	79.00	Before 2008
9	Torotorofotsy	97.64	Before 2008
10	Bemarivo	115.70	Before 2008
1	Nosy Antsoha	0.28	From 2008
2	Andreba	0.39	From 2008
3	Ampotaka Ankorabe	0.97	From 2008
4	Ankafobe	1.35	From 2008
5	Nosy Tanihely	1.80	From 2008
6	Analalava	2.29	From 2008
7	Forêt Naturel de Petriky	3.00	From 2008
8	Allées des Baobabs	3.21	From 2008
9	Mandena	4.30	From 2008
10	Analabe-Betanatanana	4.35	From 2008

Code

# Enregistrement du tableau
write_csv(desc_AP_combined, "data/derived/tableau_descriptive_AP.csv")

7.2 Statistiques descriptives du wealth index et du z-score du wealth index

Code

make_desc_wealth <- function(year){
  
  file_path <- paste0("data/derived/hh_", year, "_rural_simpler.rds")
  hh_data <- readRDS(file_path)
  
  # Wealth index en centile
  desc_centile <- hh_data %>%
    summarise(
      n_centile = sum(!is.na(wealth_centile_rural_simple)),
      mean_centile = mean(wealth_centile_rural_simple, na.rm = TRUE),
      sd_centile = sd(wealth_centile_rural_simple, na.rm = TRUE),
      median_centile = median(wealth_centile_rural_simple, na.rm = TRUE),
      min_centile = min(wealth_centile_rural_simple, na.rm = TRUE),
      max_centile = max(wealth_centile_rural_simple, na.rm = TRUE)
    ) %>%
    mutate(
      year = year,
      variable = "Wealth index en centile"
    ) %>%
    select(year, variable, everything())
  
  # Z-score Wealth index 
  desc_zscore <- hh_data %>%
    summarise(
      n_zscore = sum(!is.na(zscore_wealth)),
      mean_zscore = mean(zscore_wealth, na.rm = TRUE),
      sd_zscore = sd(zscore_wealth, na.rm = TRUE),
      median_zscore = median(zscore_wealth, na.rm = TRUE),
      min_zscore = min(zscore_wealth, na.rm = TRUE),
      max_zscore = max(zscore_wealth, na.rm = TRUE)
    ) %>%
    mutate(
      year = year,
      variable = "Z-score wealth index"
      ) %>%
    select(year, variable, everything())
  
  list(centile = desc_centile, zscore = desc_zscore)
}

years <- c(1997, 2008, 2011, 2013, 2016, 2021)

desc_list <- map(years, make_desc_wealth)

desc_centile_all <- map_dfr(desc_list, "centile") %>%
  mutate(across(where(is.numeric), ~round(.x, 2)))

desc_zscore_all <- map_dfr(desc_list, "zscore")%>%
  mutate(across(where(is.numeric), ~round(.x, 2)))

write_xlsx(
  list(
    "wealth_index_centile" = desc_centile_all, 
    "zscore_wealth_index" = desc_zscore_all
  ),
  "data/derived/descriptive_wealth_rural.xlsx"
)

kable(
  desc_centile_all,
  caption = "Statistiques descriptives - wealth index rural en centile", digits = 2)

Statistiques descriptives - wealth index rural en centile
year	variable	n_centile	mean_centile	sd_centile	median_centile	min_centile	max_centile
1997	Wealth index en centile	5124	50.32	28.92	50	1	100
2008	Wealth index en centile	13364	50.41	28.85	50	1	100
2011	Wealth index en centile	6025	50.34	28.91	50	1	100
2013	Wealth index en centile	6375	50.35	28.83	50	1	100
2016	Wealth index en centile	9295	50.47	28.85	50	1	100
2021	Wealth index en centile	15364	50.43	28.85	50	1	100

Code

kable(
  desc_zscore_all,
  caption = "Statistiques descriptives - zscore wealth index", digits = 2)

Statistiques descriptives - zscore wealth index
year	variable	n_zscore	mean_zscore	sd_zscore	median_zscore	max_zscore
1997	Z-score wealth index	5124	0.81	0.56	0.74	4.01
2008	Z-score wealth index	13364	0.80	0.57	0.71	4.66
2011	Z-score wealth index	6025	0.79	0.58	0.70	4.78
2013	Z-score wealth index	6375	0.80	0.57	0.72	4.30
2016	Z-score wealth index	9295	0.80	0.57	0.72	4.77
2021	Z-score wealth index	15364	0.80	0.57	0.72	4.40

7.3 Statistiques descriptives des variables utilisées dans le modèle

Code

library(tidyverse)
library(sf)
library(haven)

# Load data 
hr_1997_final <- readRDS("data/derived/hr_1997_final.rds")

hr_2008_final <- readRDS("data/derived/hr_2008_final.rds")

hr_2011_final <- readRDS("data/derived/hr_2011_final.rds")

hr_2013_final <- readRDS("data/derived/hr_2013_final.rds")

hr_2016_final <- readRDS("data/derived/hr_2016_final.rds")
  
hr_2021_final <- readRDS("data/derived/hr_2021_final.rds")

# Sélection des colonnes d'intérêt
vars_ut <- c("treecover_area_2000", "slope_2000", "elevation_2000", "population_count_2000", "traveltime_2000_2000", "spei_wc_1995", "spei_wc_1996", "spei_wc_1997", "spei_wc_2006", "spei_wc_2007", "spei_wc_2008","spei_wc_2019", "spei_wc_2020", "spei_wc_2021" )

# Fonction pour statistiques descriptives robustes
calc_descr_stats <- function(df, vars, year) {
  df %>% 
    select(any_of(vars)) %>% 
    mutate(across(everything(), as.numeric)) %>% 
    pivot_longer(
      everything(),
      names_to = "Variable",
      values_to = "value"
    ) %>% 
    group_by(Variable) %>% 
    summarise(
      n      = sum(!is.na(value)),
      mean   = mean(value, na.rm = TRUE),
      sd     = sd(value, na.rm = TRUE),
      min    = min(value, na.rm = TRUE),
      median = median(value, na.rm = TRUE),
      max    = max(value, na.rm = TRUE),
      .groups = "drop"
    ) %>% 
    mutate(Year = year, .before = 1)
}

years <- c(1997, 2008, 2011, 2013, 2016, 2021)
data_list <- lapply(years, function(yr){
  readRDS(glue::glue("data/derived/hr_{yr}_final.rds"))
})


descr_stats_vars <- map2_dfr(data_list, years, calc_descr_stats, vars = vars_ut) %>% 
  arrange(Variable, Year) %>% 
  mutate(across(where(is.numeric), ~round(.x, 2)))

kable(
  descr_stats_vars,
  caption = "Statistiques descriptives des covariables du modèle",
  digits = 2
)

Statistiques descriptives des covariables du modèle
Year	Variable	n	mean	sd	min	median	max
1997	elevation_2000	5124	558.86	539.42	3.01	314.74	1966.77
2008	elevation_2000	13181	597.18	530.37	2.93	435.35	1835.09
2011	elevation_2000	5997	570.49	510.53	7.47	332.11	1734.68
2013	elevation_2000	6375	571.60	515.47	9.53	329.71	1596.12
2016	elevation_2000	9295	561.30	506.07	5.53	348.85	1663.60
2021	elevation_2000	15364	602.38	519.44	5.61	460.17	2006.92
1997	population_count_2000	5124	106.14	323.19	1.79	33.09	2199.75
2008	population_count_2000	13181	77.30	248.93	1.37	27.65	2650.08
2011	population_count_2000	5997	142.82	422.06	2.31	28.86	2548.30
2013	population_count_2000	6375	124.84	375.61	1.12	27.27	2723.96
2016	population_count_2000	9295	83.71	260.86	1.16	27.70	2443.56
2021	population_count_2000	15364	88.68	304.20	0.41	24.16	2748.21
1997	slope_2000	5124	8.25	4.50	1.45	8.08	19.30
2008	slope_2000	13181	7.94	4.39	0.90	7.59	20.41
2011	slope_2000	5997	6.85	4.69	1.37	5.72	18.23
2013	slope_2000	6375	7.14	4.74	1.51	5.87	19.38
2016	slope_2000	9295	7.75	4.13	0.98	7.70	20.86
2021	slope_2000	15364	8.02	4.43	0.99	7.80	20.84
1997	spei_wc_1995	5124	-0.02	0.45	-1.18	-0.02	1.14
1997	spei_wc_1996	5124	0.90	0.69	-0.38	1.09	1.87
1997	spei_wc_1997	5124	-0.15	0.60	-1.30	-0.08	1.56
2008	spei_wc_2006	13181	0.22	0.15	-0.22	0.23	0.55
2008	spei_wc_2007	13181	0.09	0.45	-2.13	0.20	0.52
2008	spei_wc_2008	13181	0.22	0.19	-0.40	0.25	0.54
2021	spei_wc_2019	15364	-0.13	0.30	-1.15	-0.16	0.76
2021	spei_wc_2020	15364	-0.05	0.29	-0.76	-0.07	0.72
2021	spei_wc_2021	15364	-1.11	0.46	-1.91	-1.20	0.13
1997	traveltime_2000_2000	5124	247.61	190.77	14.12	204.87	1287.92
2008	traveltime_2000_2000	13181	232.09	143.89	12.31	204.77	1311.45
2011	traveltime_2000_2000	5997	198.19	137.05	14.21	179.63	949.12
2013	traveltime_2000_2000	6375	198.84	133.41	14.37	177.58	978.14
2016	traveltime_2000_2000	9295	225.11	143.74	16.35	198.95	1122.49
2021	traveltime_2000_2000	15364	238.17	154.81	13.85	212.18	1567.83
1997	treecover_area_2000	5124	18640.42	10058.05	1342.77	18447.03	31566.59
2008	treecover_area_2000	13181	16731.16	10224.05	327.77	15681.88	31566.18
2011	treecover_area_2000	5997	13857.05	10325.87	358.71	10847.27	31554.36
2013	treecover_area_2000	6375	14715.03	10354.71	146.24	12698.52	31553.77
2016	treecover_area_2000	9295	16894.43	10565.34	35.10	15765.97	31582.40
2021	treecover_area_2000	15364	16791.05	10146.62	84.97	15583.59	31547.64

7.4 Statistiques descriptives (partie matching)

Nous créons une fonction pour automatiser la création d’un résumé statistique pour chaque année.

Code

# Load data 
years <- c(1997, 2008, 2011, 2013, 2016, 2021)

summary_matched <- function(year) {
  file_path <- paste0("data/derived/data_matched_", year, ".rds")
  df <- readRDS(file_path) %>% 
    mutate(across(where(is.numeric), as.numeric))

  df %>% 
    pivot_longer(
      where(is.numeric),
      names_to = "Variable",
      values_to = "value"
    ) %>% 
    group_by(Variable) %>% 
    summarise(
      n      = sum(!is.na(value)),
      mean   = mean(value, na.rm = TRUE),
      sd     = sd(value, na.rm = TRUE),
      min    = min(value, na.rm = TRUE),
      median = median(value, na.rm = TRUE),
      max    = max(value, na.rm = TRUE),
      .groups = "drop"
    ) %>% 
    mutate(Year = year, .before = 1)
}

descr_stats_match <- map_dfr(years, summary_matched) %>% 
  arrange(Variable, Year) %>% 
  mutate(across(where(is.numeric), ~round(.x, 2)))

kable(
  descr_stats_match,
  caption = "Statistiques descriptives des variables dans les échantillons appariés",
  digits = 2
)

Statistiques descriptives des variables dans les échantillons appariés
Year	Variable	n	mean	sd	min	median	max
1997	DHSYEAR	1930	1997.00	0.00	1997.00	1997.00	1997.00
2008	DHSYEAR	4902	2008.00	0.00	2008.00	2008.00	2008.00
2011	DHSYEAR	1860	2011.00	0.00	2011.00	2011.00	2011.00
2013	DHSYEAR	3078	2013.00	0.00	2013.00	2013.00	2013.00
2016	DHSYEAR	3370	2016.00	0.00	2016.00	2016.00	2016.00
2021	DHSYEAR	6784	2021.00	0.00	2021.00	2021.00	2021.00
1997	STATUS_YR	965	2014.80	1.02	2009.00	2015.00	2015.00
2008	STATUS_YR	2451	2014.52	1.39	2009.00	2015.00	2015.00
2011	STATUS_YR	930	2014.69	1.15	2010.00	2015.00	2015.00
2013	STATUS_YR	1539	2014.60	1.19	2010.00	2015.00	2015.00
2016	STATUS_YR	1685	2014.81	1.05	2010.00	2015.00	2017.00
2021	STATUS_YR	3392	2014.46	1.54	2010.00	2015.00	2017.00
1997	WDPAID	965	285717732.96	277766602.67	1299.00	555542728.00	555697917.00
2008	WDPAID	2451	307040577.16	276267047.27	1299.00	555547960.00	555697917.00
2011	WDPAID	930	426639369.17	234644631.79	1299.00	555549452.00	555785986.00
2013	WDPAID	1539	352837539.73	267485964.60	1299.00	555549452.00	555785986.00
2016	WDPAID	1685	324916381.87	273784180.32	1299.00	555549450.00	555697916.00
2021	WDPAID	3392	364405870.81	263955383.79	1299.00	555697871.00	555697918.00
1997	dist_km	965	3.90	3.03	0.00	3.78	8.91
2008	dist_km	2451	4.64	3.06	0.00	5.23	10.01
2011	dist_km	930	4.31	3.38	0.00	5.50	9.93
2013	dist_km	1539	4.50	3.41	0.00	4.37	9.89
2016	dist_km	1685	4.56	3.29	0.00	4.86	9.68
2021	dist_km	3392	4.83	3.33	0.00	5.18	9.95
1997	elevation_2000	1930	403.85	443.83	3.01	207.42	1892.28
2008	elevation_2000	4902	404.32	470.78	2.93	195.40	1800.69
2011	elevation_2000	1860	380.94	449.69	7.47	167.05	1555.08
2013	elevation_2000	3078	315.34	404.44	9.53	157.55	1596.12
2016	elevation_2000	3370	377.12	450.11	5.53	164.99	1663.60
2021	elevation_2000	6784	405.23	472.56	5.61	188.91	1824.29
1997	hv001	1930	178.41	59.96	64.00	181.00	270.00
2008	hv001	4902	381.47	147.28	62.00	377.00	600.00
2011	hv001	1860	162.13	71.48	12.00	189.00	268.00
2013	hv001	3078	171.61	70.95	13.00	186.00	284.00
2016	hv001	3370	232.23	94.47	24.00	239.00	375.00
2021	hv001	6784	404.87	162.59	52.00	409.00	655.00
1997	hv002	1930	19.75	14.10	1.00	17.00	60.00
2008	hv002	4902	82.31	63.03	1.00	69.00	446.00
2011	hv002	1860	87.38	62.43	1.00	77.00	296.00
2013	hv002	3078	77.42	51.46	3.00	71.00	274.00
2016	hv002	3370	116.28	535.16	1.00	77.00	9040.00
2021	hv002	6784	84.37	61.93	1.00	75.00	1146.00
1997	hv005	1930	993831.26	258108.51	602637.00	1070160.00	1490113.00
2008	hv005	4902	1092060.84	632228.03	146947.00	932684.00	2806127.00
2011	hv005	1860	986568.93	960310.64	47981.00	451523.00	3962731.00
2013	hv005	3078	1266857.67	951224.57	146497.00	1346608.00	4111918.00
2016	hv005	3370	1077199.12	667117.77	120564.00	941891.00	3421986.00
2021	hv005	6784	1067029.08	564203.69	98180.00	950214.50	3858947.00
1997	hv025	1930	2.00	0.00	2.00	2.00	2.00
2008	hv025	4902	2.00	0.00	2.00	2.00	2.00
2011	hv025	1860	2.00	0.00	2.00	2.00	2.00
2013	hv025	3078	2.00	0.00	2.00	2.00	2.00
2016	hv025	3370	2.00	0.00	2.00	2.00	2.00
2021	hv025	6784	2.00	0.00	2.00	2.00	2.00
1997	hv219	1930	1.23	0.42	1.00	1.00	2.00
2008	hv219	4902	1.22	0.41	1.00	1.00	2.00
2011	hv219	1860	1.27	0.44	1.00	1.00	2.00
2013	hv219	3078	1.26	0.44	1.00	1.00	2.00
2016	hv219	3370	1.32	0.47	1.00	1.00	2.00
2021	hv219	6784	1.23	0.42	1.00	1.00	2.00
1997	hv220	1929	43.11	16.21	10.00	40.00	98.00
2008	hv220	4902	42.57	15.70	13.00	40.00	99.00
2011	hv220	1860	43.62	16.24	13.00	41.00	98.00
2013	hv220	3078	43.75	16.77	14.00	42.00	98.00
2016	hv220	3370	42.67	16.00	13.00	40.00	98.00
2021	hv220	6784	42.66	15.74	15.00	40.00	95.00
1997	hv271	1930	-0.48	0.39	-1.09	-0.56	1.87
2008	hv271	4902	-46649.50	55805.66	-117457.00	-65227.50	257881.00
2011	hv271	1860	-49077.32	56852.63	-130632.00	-67778.50	303130.00
2013	hv271	3078	-53964.28	43195.87	-106583.00	-67441.00	221809.00
2016	hv271	3370	-34437.66	66445.81	-134245.00	-56575.50	417411.00
2021	hv271	6784	-39468.04	62276.79	-119182.00	-56674.50	407214.00
1997	population_count_2000	1930	37.88	32.12	2.33	26.81	155.18
2008	population_count_2000	4902	34.71	29.14	1.39	27.65	178.49
2011	population_count_2000	1860	42.69	27.66	5.49	40.31	130.01
2013	population_count_2000	3078	31.45	23.80	1.12	25.97	136.35
2016	population_count_2000	3370	33.42	27.25	1.16	25.31	119.65
2021	population_count_2000	6784	32.57	32.75	0.41	24.16	232.04
1997	slope_2000	1930	8.66	5.00	1.45	9.40	16.69
2008	slope_2000	4902	8.31	5.16	0.90	7.57	17.67
2011	slope_2000	1860	6.18	5.23	1.37	3.18	18.23
2013	slope_2000	3078	6.90	5.22	1.58	4.58	18.72
2016	slope_2000	3370	7.42	4.88	0.98	5.63	17.88
2021	slope_2000	6784	8.48	4.98	0.99	8.89	20.84
1997	spei_wc_1995	1930	0.06	0.46	-0.78	0.07	1.14
1997	spei_wc_1996	1930	0.79	0.62	-0.29	0.81	1.79
1997	spei_wc_1997	1930	-0.24	0.58	-1.30	-0.20	1.56
2008	spei_wc_2006	4902	0.23	0.17	-0.22	0.27	0.55
2008	spei_wc_2007	4902	-0.02	0.66	-2.13	0.23	0.52
2008	spei_wc_2008	4902	0.21	0.24	-0.40	0.29	0.54
2011	spei_wc_2009	1860	0.33	0.09	-0.06	0.33	0.60
2011	spei_wc_2010	1860	0.10	0.18	-0.35	0.09	0.92
2011	spei_wc_2011	1860	-0.36	0.36	-1.12	-0.23	0.65
2013	spei_wc_2011	3078	-0.33	0.36	-1.13	-0.21	0.67
2013	spei_wc_2012	3078	-0.37	0.31	-0.91	-0.33	0.41
2013	spei_wc_2013	3078	0.52	1.17	-1.83	0.64	1.96
2016	spei_wc_2014	3370	0.07	0.98	-1.42	0.13	1.77
2016	spei_wc_2015	3370	0.20	0.74	-0.98	0.35	2.01
2016	spei_wc_2016	3370	-0.48	0.56	-1.21	-0.55	0.99
2021	spei_wc_2019	6784	-0.17	0.34	-1.15	-0.24	0.76
2021	spei_wc_2020	6784	-0.09	0.32	-0.74	-0.10	0.71
2021	spei_wc_2021	6784	-0.96	0.52	-1.91	-1.15	0.13
1997	traveltime_2000_2000	1930	221.34	110.83	46.75	204.87	692.19
2008	traveltime_2000_2000	4902	238.38	128.46	29.54	208.11	614.27
2011	traveltime_2000_2000	1860	154.80	94.05	36.21	135.78	511.82
2013	traveltime_2000_2000	3078	218.55	145.24	36.85	190.12	978.14
2016	traveltime_2000_2000	3370	235.92	127.02	44.72	216.92	612.61
2021	traveltime_2000_2000	6784	257.36	146.30	42.24	221.96	1428.54
1997	treatment	1930	0.50	0.50	0.00	0.50	1.00
2008	treatment	4902	0.50	0.50	0.00	0.50	1.00
2011	treatment	1860	0.50	0.50	0.00	0.50	1.00
2013	treatment	3078	0.50	0.50	0.00	0.50	1.00
2016	treatment	3370	0.50	0.50	0.00	0.50	1.00
2021	treatment	6784	0.50	0.50	0.00	0.50	1.00
1997	treecover_area_2000	1930	22346.44	9084.92	1342.77	24936.56	31566.59
2008	treecover_area_2000	4902	21427.72	9296.08	327.77	23493.47	31566.18
2011	treecover_area_2000	1860	14741.10	11246.84	358.71	11327.65	31541.12
2013	treecover_area_2000	3078	18492.66	10869.12	146.24	20201.42	31553.77
2016	treecover_area_2000	3370	21634.03	9026.54	35.10	24420.51	31582.40
2021	treecover_area_2000	6784	22541.23	8960.72	952.56	24825.72	31547.64
1997	wealth_centile_rural_simple	1930	48.43	26.68	1.00	49.00	100.00
2008	wealth_centile_rural_simple	4902	47.76	28.17	1.00	46.00	100.00
2011	wealth_centile_rural_simple	1860	44.66	27.85	1.00	42.00	100.00
2013	wealth_centile_rural_simple	3078	45.16	26.16	1.00	45.00	99.00
2016	wealth_centile_rural_simple	3370	47.58	28.04	1.00	46.00	100.00
2021	wealth_centile_rural_simple	6784	50.02	27.79	1.00	50.00	100.00
1997	wealth_centile_rural_weighted	1930	47.43	26.51	1.00	49.00	100.00
2008	wealth_centile_rural_weighted	4902	42.76	27.70	1.00	39.00	100.00
2011	wealth_centile_rural_weighted	1860	39.90	29.24	1.00	34.00	100.00
2013	wealth_centile_rural_weighted	3078	41.92	26.81	1.00	39.00	99.00
2016	wealth_centile_rural_weighted	3370	44.60	27.32	1.00	43.00	100.00
2021	wealth_centile_rural_weighted	6784	46.03	27.32	1.00	45.00	100.00
1997	weights	1930	1.00	0.00	1.00	1.00	1.00
2008	weights	4902	1.00	0.00	1.00	1.00	1.00
2011	weights	1860	1.00	0.00	1.00	1.00	1.00
2013	weights	3078	1.00	0.00	1.00	1.00	1.00
2016	weights	3370	1.00	0.00	1.00	1.00	1.00
2021	weights	6784	1.00	0.00	1.00	1.00	1.00
1997	zscore_wealth	1930	0.81	0.55	0.00	0.74	3.81
2008	zscore_wealth	4902	0.80	0.57	0.00	0.72	4.50
2011	zscore_wealth	1860	0.81	0.56	0.00	0.73	3.22
2013	zscore_wealth	3078	0.80	0.57	0.00	0.73	3.94
2016	zscore_wealth	3370	0.80	0.58	0.00	0.71	4.77
2021	zscore_wealth	6784	0.80	0.57	0.00	0.72	3.93

7.1 Statistiques descriptive des aires protégées

7.2 Statistiques descriptives du wealth index et du z-score du wealth index

7.3 Statistiques descriptives des variables utilisées dans le modèle

7.4 Statistiques descriptives (partie matching)

7.5 Statistiques descriptives (partie estimation)