One longstanding debate in the study of international relations concerns the question of whether individual political leaders can make a difference. Some emphasize that leaders with different ideologies and personalities can significantly affect the course of a nation. Others argue that political leaders are severely constrained by historical and institutional forces. Did individuals like Hitler, Mao, Roosevelt, and Churchill make a big difference? The difficulty of empirically testing these arguments stems from the fact that the change of leadership is not random and there are many confounding factors to be adjusted for. In this exercise, we consider a natural experiment in which the success or failure of assassination attempts is assumed to be essentially random.7 Each observation of the CSV data set leaders.csv contains information about an assassination attempt. Table 2.8 presents the names and descriptions of variables in this leader assassination data set. The polity variable represents the so-called polity score from the Polity Project. The Polity Project systematically documents and quantifies the regime types of all countries in the world from 1800. The polity score is a 21-point scale ranging from ???10 (hereditary monarchy) to 10 (consolidated democracy). The result variable is a 10-category factor variable describing the result of each assassination attempt.
Load the data from github
library("RCurl")## Warning: package 'RCurl' was built under R version 3.3.3## Loading required package: bitopsx <- getURL("https://raw.githubusercontent.com/kosukeimai/qss/master/CAUSALITY/leaders.csv")
leaders <- read.csv(text = x)Explore the dataset
head(leaders)## year country leadername age politybefore polityafter
## 1 1929 Afghanistan Habibullah Ghazi 39 -6 -6.000000
## 2 1933 Afghanistan Nadir Shah 53 -6 -7.333333
## 3 1934 Afghanistan Hashim Khan 50 -6 -8.000000
## 4 1924 Albania Zogu 29 0 -9.000000
## 5 1931 Albania Zogu 36 -9 -9.000000
## 6 1968 Algeria Boumedienne 41 -9 -9.000000
## interwarbefore interwarafter civilwarbefore civilwarafter
## 1 0 0 1 0
## 2 0 0 0 0
## 3 0 0 0 0
## 4 0 0 0 0
## 5 0 0 0 0
## 6 0 0 0 0
## result
## 1 not wounded
## 2 dies within a day after the attack
## 3 survives, whether wounded unknown
## 4 wounded lightly
## 5 not wounded
## 6 wounded lightlysummary(leaders)## year country leadername age
## Min. :1878 Japan : 11 Mussolini : 6 Min. :18.00
## 1st Qu.:1920 Mexico : 11 Alexander II : 4 1st Qu.:45.00
## Median :1949 France : 10 De Gaulle : 4 Median :52.50
## Mean :1945 Russia : 10 Alexander III: 3 Mean :53.52
## 3rd Qu.:1972 United States: 8 Amin : 3 3rd Qu.:61.75
## Max. :2001 Guatemala : 7 Carlos I : 3 Max. :81.00
## (Other) :193 (Other) :227
## politybefore polityafter interwarbefore interwarafter
## Min. :-10.000 Min. :-10.000 Min. :0.000 Min. :0.000
## 1st Qu.: -7.000 1st Qu.: -7.000 1st Qu.:0.000 1st Qu.:0.000
## Median : -3.000 Median : -3.167 Median :0.000 Median :0.000
## Mean : -1.519 Mean : -1.650 Mean :0.188 Mean :0.148
## 3rd Qu.: 4.000 3rd Qu.: 3.917 3rd Qu.:0.000 3rd Qu.:0.000
## Max. : 10.000 Max. : 10.000 Max. :1.000 Max. :1.000
##
## civilwarbefore civilwarafter
## Min. :0.000 Min. :0.000
## 1st Qu.:0.000 1st Qu.:0.000
## Median :0.000 Median :0.000
## Mean :0.216 Mean :0.184
## 3rd Qu.:0.000 3rd Qu.:0.000
## Max. :1.000 Max. :1.000
##
## result
## not wounded :96
## dies within a day after the attack :46
## plot stopped :40
## wounded lightly :25
## hospitalization but no permanent disability:20
## survives, whether wounded unknown :14
## (Other) : 91. How many assassination attempts are recorded in the data? How many countries experience at least one leader assassination attempt? (The unique() function which returns a set of unique values from the input vector, may be useful here.) What is the average number of such attempts (per year) among these countries?
Use unique() to gee unique values in the country variables, and length()
length(unique(leaders$country))## [1] 88We need the average number of assassination attempts among these countries for each year, which then needs to be averaged across years
mean(tapply(leaders$country, leaders$year, length))## [1] 2.45098This is the mean of the mean number of countries for each year.
2. Create a new binary variable named success that is equal to 1 if a leader dies from the attack and 0 if the leader survives. Store this new variable as part of the original data frame. What is the overall success rate of leader assassination? Does the result speak to the validity of the assumption that the success of assassination attempts is randomly determined?
table(leaders$result)##
## dies between a day and a week
## 2
## dies between a week and a month
## 2
## dies within a day after the attack
## 46
## dies, timing unknown
## 4
## hospitalization but no permanent disability
## 20
## not wounded
## 96
## plot stopped
## 40
## survives but wounded severely
## 1
## survives, whether wounded unknown
## 14
## wounded lightly
## 25levels(leaders$result)## [1] "dies between a day and a week"
## [2] "dies between a week and a month"
## [3] "dies within a day after the attack"
## [4] "dies, timing unknown"
## [5] "hospitalization but no permanent disability"
## [6] "not wounded"
## [7] "plot stopped"
## [8] "survives but wounded severely"
## [9] "survives, whether wounded unknown"
## [10] "wounded lightly"leaders$success <- ifelse(leaders$result == "dies between a day and a week" |
leaders$result == "dies between a week and a month" |
leaders$result == "dies within a day after the attack" |
leaders$result == "dies, timing unknown", 1, 0)
prop.table(table(leaders$result, leaders$success), 1)##
## 0 1
## dies between a day and a week 0 1
## dies between a week and a month 0 1
## dies within a day after the attack 0 1
## dies, timing unknown 0 1
## hospitalization but no permanent disability 1 0
## not wounded 1 0
## plot stopped 1 0
## survives but wounded severely 1 0
## survives, whether wounded unknown 1 0
## wounded lightly 1 0mean(leaders$success) # 22% of assissination attempts end in death## [1] 0.2163. Investigate whether the average polity score over three years prior to an assassination attempt differs on average between successful and failed attempts. Also, examine whether there is any difference in the age of targeted leaders between successful and failed attempts. Briefly interpret the results in light of the validity of the aforem4. Repeat the same analysis as in the previous question, but this time using the country’s experience of civil and international war. Create a new binary variable in the data frame called warbefore. Code the variable such that it is equal to 1 if a country is in either civil or international war during the three years prior to an assassination attempt. Provide a brief interpretation of the result.entioned assumption.
summary(leaders)## year country leadername age
## Min. :1878 Japan : 11 Mussolini : 6 Min. :18.00
## 1st Qu.:1920 Mexico : 11 Alexander II : 4 1st Qu.:45.00
## Median :1949 France : 10 De Gaulle : 4 Median :52.50
## Mean :1945 Russia : 10 Alexander III: 3 Mean :53.52
## 3rd Qu.:1972 United States: 8 Amin : 3 3rd Qu.:61.75
## Max. :2001 Guatemala : 7 Carlos I : 3 Max. :81.00
## (Other) :193 (Other) :227
## politybefore polityafter interwarbefore interwarafter
## Min. :-10.000 Min. :-10.000 Min. :0.000 Min. :0.000
## 1st Qu.: -7.000 1st Qu.: -7.000 1st Qu.:0.000 1st Qu.:0.000
## Median : -3.000 Median : -3.167 Median :0.000 Median :0.000
## Mean : -1.519 Mean : -1.650 Mean :0.188 Mean :0.148
## 3rd Qu.: 4.000 3rd Qu.: 3.917 3rd Qu.:0.000 3rd Qu.:0.000
## Max. : 10.000 Max. : 10.000 Max. :1.000 Max. :1.000
##
## civilwarbefore civilwarafter
## Min. :0.000 Min. :0.000
## 1st Qu.:0.000 1st Qu.:0.000
## Median :0.000 Median :0.000
## Mean :0.216 Mean :0.184
## 3rd Qu.:0.000 3rd Qu.:0.000
## Max. :1.000 Max. :1.000
##
## result success
## not wounded :96 Min. :0.000
## dies within a day after the attack :46 1st Qu.:0.000
## plot stopped :40 Median :0.000
## wounded lightly :25 Mean :0.216
## hospitalization but no permanent disability:20 3rd Qu.:0.000
## survives, whether wounded unknown :14 Max. :1.000
## (Other) : 9Three years prior to an attempt (politybefore, polityafter) - already derived
mean(leaders$politybefore)## [1] -1.518667mean(leaders$polityafter)## [1] -1.65These are fairly similar - without taking account of whether attempt was successful or not
tapply(leaders$politybefore, leaders$success, mean)## 0 1
## -1.7431973 -0.7037037The polity score is higher in countries where a successful assassination attempt occurs (more democratic)
tapply(leaders$age, leaders$success, mean)## 0 1
## 52.71429 56.46296Successful assassination attempts are associated with older leaders
4. Repeat the same analysis as in the previous question, but this time using the country’s experience of civil and international war. Create a new binary variable in the data frame called warbefore. Code the variable such that it is equal to 1 if a country is in either civil or international war during the three years prior to an assassination attempt. Provide a brief interpretation of the result.
table(leaders$interwarbefore)##
## 0 1
## 203 47table(leaders$civilwarbefore)##
## 0 1
## 196 54leaders$warbefore <- ifelse(leaders$interwarbefore == 1 | leaders$civilwarbefore == 1, 1, 0)
table(leaders$interwarbefore, leaders$civilwarbefore)##
## 0 1
## 0 158 45
## 1 38 945 + 38 + 9## [1] 92table(leaders$warbefore)##
## 0 1
## 158 92Just double checking
mean(leaders$warbefore)## [1] 0.368tapply(leaders$warbefore, leaders$success, mean)## 0 1
## 0.3724490 0.3518519Not a huge difference
Example in the book chapter does it differently (see below)
mean(leaders$warbefore[leaders$success == 1])## [1] 0.3518519mean(leaders$warbefore[leaders$success == 0])## [1] 0.3724495. Does successful leader assassination cause democratization? Does successful leader assassination lead countries to war? When analyzing these data, be sure to state your assumptions and provide a brief interpretation of the results.
This has got to be the difference in differences model
polity.change.treat <- mean(leaders$polityafter[leaders$success == 1]) - mean(leaders$politybefore[leaders$success == 1])
polity.change.control <- mean(leaders$polityafter[leaders$success == 0]) - mean(leaders$politybefore[leaders$success == 0])
polity.change.treat - polity.change.control## [1] 0.09271857Really very minor change - probably not significant - very slightly more democratic
mean(leaders$polityafter[leaders$success == 1]) - mean(leaders$polityafter[leaders$success == 0])## [1] 1.132212Naive comparison of polity afterwards would suggest that the assassination had made a difference
Create variable for warafter
leaders$warafter <- ifelse(leaders$interwarafter == 1 |
leaders$civilwarafter == 1, 1, 0)Compare war before to war after among successful and unsuccessful
diff.war.succ <- mean(leaders$warafter[leaders$success == 1]) -
mean(leaders$warbefore[leaders$success == 1])
diff.war.unsucc <-mean(leaders$warafter[leaders$success == 0]) -
mean(leaders$warbefore[leaders$success == 0])
diff.war.succ - diff.war.unsucc # difference in differences## [1] -0.07161754Not much difference