library(igoR)
# Load helper packages.
library(ggplot2)
library(dplyr)Get started with the igoR package
This vignette provides a visual introduction to igoR.
The examples are based on Pevehouse et al. (2020). For more information about the IGO data sets and additional downloads, see Intergovernmental Organizations (v3).
The dyadic data set is not included in the package due to its size (~500 MB in Stata .dta format), but igo_dyadic() provides comparable joint membership results.
Definitions
From Pevehouse et al. (2019):
What is an IGO?
The original data set defines an Intergovernmental Organization (IGO) using the following criteria:
- An IGO must consist of at least three members of the COW-defined state system.
- An IGO must hold regular plenary sessions at least once every ten years.
- An IGO must possess a permanent secretariat and corresponding headquarters.
When does an IGO actually begin?
The data set begins to code an IGO by identifying the first year in which the organization functions. In some cases, individual members are listed by year of accession or signature.
When does an IGO die?
Version 3.0 of the IGO data set uses the following criteria:
- An organization is considered terminated when its context is described with one of the following terms: replaced, succeeded, superseded, integrated, merged or dies.
Analysis
This section provides a short analysis based on figures in Pevehouse et al. (2020).
Initial setup
First, create a custom ggplot2::theme() named theme_igor. The figures in this vignette use this theme.
theme_igor <- theme(
axis.title = element_blank(),
axis.line.x.bottom = element_line("black"),
axis.line.y.left = element_line("black"),
axis.text = element_text(color = "black", family = "sans"),
axis.text.y.left = element_text(angle = 90, hjust = 0.5),
legend.position = "bottom",
legend.title = element_blank(),
legend.key = element_blank(),
legend.key.width = unit(2, "cm"),
legend.text = element_text(family = "sans", size = 11.5),
legend.box.background = element_rect(color = "black", linewidth = 1),
legend.spacing = unit(1.2 / 100, "npc"),
plot.background = element_rect("grey90"),
plot.margin = unit(rep(0.5, 4), "cm"),
panel.background = element_rect("white"),
panel.grid = element_blank(),
panel.border = element_rect(fill = NA, colour = "grey90"),
panel.grid.major.y = element_line("grey90")
)IGOs overview
The following code counts IGOs and states included in the package. The available years are 1816 to 2014.
# Summarize values by year.
igos_by_year <- igo_year_format3 %>%
group_by(year) %>%
summarise(value = n(), .groups = "keep") %>%
mutate(variable = "Total IGOs")
countries_by_year <- state_year_format3 %>%
group_by(year) %>%
summarise(value = n(), .groups = "keep") %>%
mutate(variable = "Number of COW states")
all_by_year <- igos_by_year %>%
bind_rows(countries_by_year) %>%
# Label the plot.
mutate(
variable = factor(
variable,
levels = c("Total IGOs", "Number of COW states")
)
)
# Plot the results.
ggplot(all_by_year, aes(x = year, y = value)) +
geom_line(color = "black", aes(linetype = variable)) +
scale_x_continuous(limits = c(1800, 2014)) +
scale_linetype_manual(values = c("solid", "dashed")) +
geom_vline(xintercept = c(1945, 1989)) +
ylim(0, 400) +
theme_igor
IGO births and deaths
This plot shows how many IGOs started or ended in each year.
# Summarize IGO starts and endings by year.
df <- igo_search()
births <- df %>%
mutate(year = sdate) %>%
group_by(year) %>%
summarise(value = n(), .groups = "keep") %>%
mutate(variable = "IGO Births")
deads <- df %>%
mutate(year = deaddate) %>%
group_by(year) %>%
summarise(value = n(), .groups = "keep") %>%
mutate(variable = "IGO Deaths")
births_and_deads <- births %>%
bind_rows(deads) %>%
filter(!is.na(year))
# Plot the results.
ggplot(births_and_deads, aes(x = year, y = value)) +
geom_line(color = "black", aes(linetype = variable)) +
scale_linetype_manual(values = c("solid", "dashed")) +
scale_x_continuous(
limits = c(1815, 2015),
breaks = seq(1815, 2015, by = 25)
) +
ylim(0, 15) +
theme_igor
IGOs across regions
This plot shows the number of IGOs by region. The region definitions are based on Pevehouse et al. (2020) and the complementary replication data set (PRIO 2020).
IGOs across regions: codes
# Cross-regional and universal codes are not included.
asia <- c(
550,
560,
570,
580,
590,
600,
610,
640,
650,
660,
670,
725,
750,
825,
1030,
1345,
1400,
1530,
1532,
2300,
2770,
3185,
3330,
3560,
3930,
4115,
4150,
4160,
4170,
4190,
4200,
4220,
4265,
4440
)
middle_east <- c(
370,
380,
390,
400,
410,
420,
430,
440,
450,
460,
470,
490,
500,
510,
520,
1110,
1410,
1990,
2000,
2220,
3450,
3800,
4140,
4270,
4380
)
europe <- c(
20,
300,
780,
800,
832,
840,
860,
1020,
1050,
1070,
1080,
1125,
1140,
1390,
1420,
1440,
1563,
1565,
1580,
1585,
1590,
1600,
1610,
1620,
1630,
1640,
1645,
1653,
1660,
1670,
1675,
1680,
1690,
1700,
1710,
1715,
1720,
1730,
1740,
1750,
1760,
1770,
1780,
1790,
1800,
1810,
1820,
1830,
1930,
1970,
1980,
2310,
2325,
2345,
2440,
2450,
2550,
2575,
2610,
2650,
2705,
2890,
2972,
3010,
3095,
3230,
3290,
3360,
3485,
3505,
3585,
3590,
3600,
3610,
3620,
3630,
3640,
3650,
3655,
3660,
3665,
3762,
3810,
3855,
3860,
3910,
4000,
4350,
4450,
4460,
4510,
4520,
4540
)
africa <- c(
30,
40,
50,
60,
80,
90,
100,
110,
115,
120,
125,
130,
140,
150,
155,
160,
170,
180,
190,
200,
210,
225,
240,
250,
260,
280,
290,
690,
700,
710,
940,
1060,
1150,
1170,
1260,
1290,
1310,
1320,
1330,
1340,
1355,
1430,
1450,
1460,
1470,
1475,
1480,
1500,
1510,
1520,
1870,
2080,
2090,
2230,
2330,
2795,
3300,
3310,
3470,
3480,
3510,
3520,
3570,
3740,
3760,
3761,
3790,
3820,
3875,
3905,
3970,
4010,
4030,
4050,
4055,
4080,
4110,
4120,
4130,
4230,
4240,
4250,
4251,
4340,
4365,
4480,
4485,
4490,
4500,
4501,
4503
)
americas <- c(
310,
320,
330,
340,
720,
760,
815,
875,
880,
890,
900,
910,
912,
913,
920,
950,
970,
980,
990,
1000,
1010,
1095,
1130,
1486,
1489,
1490,
1860,
1890,
1920,
1950,
2070,
2110,
2120,
2130,
2140,
2150,
2160,
2170,
2175,
2180,
2190,
2200,
2203,
2206,
2210,
2260,
2340,
2490,
2560,
2980,
3060,
3340,
3370,
3380,
3390,
3400,
3410,
3420,
3428,
3430,
3670,
3680,
3812,
3830,
3880,
3890,
3900,
3925,
3980,
4070,
4100,
4260,
4280,
4370
)# `africa`, `americas`, `asia`, `europe` and `middle_east` were created in the
# previous chunk, which is collapsed for readability.
regions <- igo_search() %>%
mutate(
region = case_when(
ionum %in% africa ~ "Africa",
ionum %in% americas ~ "Americas",
ionum %in% asia ~ "Asia",
ionum %in% europe ~ "Europe",
ionum %in% middle_east ~ "Middle East",
TRUE ~ NA
)
) %>%
select(ioname, region)After creating a data frame with the regions, classify IGOs by region.
# The `regions` data set was created in the previous chunk.
# Select all IGOs.
alligos <- igo_year_format3 %>%
select(ioname, year)
regionsum <- alligos %>%
left_join(regions) %>%
group_by(year, region) %>%
summarise(value = n(), .groups = "keep") %>%
filter(!is.na(region)) %>%
# Prepare for plotting.
mutate(
region = factor(
region,
levels = c(
"Asia",
"Europe",
"Africa",
"Americas",
"Middle East"
)
)
)
# Plot the results.
ggplot(regionsum, aes(x = year, y = value)) +
geom_line(color = "black", aes(linetype = region)) +
scale_linetype_manual(
values = c("solid", "dashed", "dotted", "dotdash", "longdash")
) +
guides(linetype = guide_legend(ncol = 2, byrow = TRUE)) +
ylim(0, 80) +
scale_x_continuous(
limits = c(1815, 2015),
breaks = seq(1815, 2015, by = 25)
) +
theme_igor
Selected states in Asia
This plot shows the number of full memberships for five states in Asia: India, China, Pakistan, Indonesia and Bangladesh.
asia5_cntries <- c("China", "India", "Pakistan", "Indonesia", "Bangladesh")
# Use five states in Asia.
asia5_igos <- igo_state_membership(
state = asia5_cntries,
year = 1865:2014,
status = "Full Membership"
)
asia5 <- asia5_igos %>%
group_by(statenme, year) %>%
summarise(values = n(), .groups = "keep") %>%
mutate(statenme = factor(statenme, levels = asia5_cntries))
# Plot the results.
ggplot(asia5, aes(x = year, y = values)) +
geom_line(color = "black", aes(linetype = statenme)) +
scale_linetype_manual(
values = c("solid", "dashed", "dotted", "dotdash", "longdash")
) +
guides(linetype = guide_legend(ncol = 3, byrow = TRUE)) +
theme(
axis.title.y.left = element_text(
family = "sans",
size = 12,
margin = margin(r = 6)
)
) +
scale_x_continuous(
limits = c(1865, 2015),
breaks = seq(1865, 2015, by = 25)
) +
scale_y_continuous(
"Number of memberships",
breaks = seq(0, 100, 20),
limits = c(0, 100)
) +
theme_igor
Shared memberships
The final plot counts shared full memberships between Spain and four selected states.
selected_countries <- c("France", "Morocco", "China", "USA")
spain_selected <- igo_dyadic("Spain", selected_countries)
# Compute the number of shared memberships.
spain_selected <- spain_selected %>%
rowwise() %>%
mutate(values = sum(c_across(aaaid:wassen) == 1))
# Plot the results.
ggplot(spain_selected, aes(x = year, y = values)) +
geom_line(color = "black", aes(linetype = statenme2)) +
scale_linetype_manual(values = c("solid", "dashed", "dotted", "dotdash")) +
guides(linetype = guide_legend(ncol = 2, byrow = TRUE)) +
theme(
axis.title.y.left = element_text(
family = "sans",
size = 10,
margin = margin(r = 6)
)
) +
scale_x_continuous(
limits = c(1815, 2015),
breaks = seq(1815, 2015, by = 25)
) +
scale_y_continuous(
"Number of memberships",
breaks = seq(0, 110, 20),
limits = c(0, 110)
) +
theme_igor +
geom_vline(xintercept = 1939, alpha = 0.2) +
annotate("label", x = 1938, y = 60, size = 3, label = "Spanish \nCivil War") +
geom_vline(xintercept = 1978, alpha = 0.2) +
annotate(
"label",
x = 1970,
y = 100,
size = 3,
label = "Constitution \nof Spain"
)
References
Pevehouse, Jon C. W., Timothy Nordstrom, Roseanne W. McManus, and Anne Spencer Jamison. 2020. “Tracking Organizations in the World: The Correlates of War IGO Version 3.0 Datasets.” Journal of Peace Research 57 (3): 492–503. https://doi.org/10.1177/0022343319881175.
Pevehouse, Jon, Roseanne McManus, and Timothy Nordstrom. 2019. Codebook for Correlates of War 3 International Governmental Organizations Data Set Version 3.0.https://correlatesofwar.org/wp-content/uploads/IGO-Codebook_v3_short-copy.pdf .
PRIO. 2020. Replication Datasets: Journal of Peace Research. Peace Research Institute Oslo; Online. https://www.prio.org/journals/jpr/replicationdata.