setwd("C:/Users/srish/Documents/ASP Thesis")
library(tidyverse)
## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr 1.1.1 ✔ readr 2.1.4
## ✔ forcats 1.0.0 ✔ stringr 1.5.0
## ✔ ggplot2 3.4.1 ✔ tibble 3.2.1
## ✔ lubridate 1.9.2 ✔ tidyr 1.3.0
## ✔ purrr 1.0.1
## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errors
library(dplyr)
library(ggplot2)
library(car)
## Loading required package: carData
##
## Attaching package: 'car'
##
## The following object is masked from 'package:dplyr':
##
## recode
##
## The following object is masked from 'package:purrr':
##
## some
library(dunn.test)
library(WRS2)
library(stargazer)
##
## Please cite as:
##
## Hlavac, Marek (2022). stargazer: Well-Formatted Regression and Summary Statistics Tables.
## R package version 5.2.3. https://CRAN.R-project.org/package=stargazer
## cleaning the data
ds<-read.csv("merged.csv")
ds$state <- trimws(ds$state, which = "right")
ds$state_party <- trimws(ds$state_party, which = "right")
ds$central_party <- trimws(ds$central_party, which = "right")
ds$pr[is.na(ds$pr)] = 0
ds$scaled[is.na(ds$scaled)] = 4
## creating misalign due to data entry error
ds<-ds%>%
filter(X_merge == "Matched (3)")%>%
mutate(misalign = case_when(
scaled == 3 ~ 0,
scaled == 4 ~ 0,
scaled == 1 ~ 1,
scaled == 2 ~ 1,
))
#### analysing one sided violence by the central govt against only civilians
state_only<- ds%>%
filter(type_of_violence == 3 & side_a == "Government of India")
view(state_only)
table(state_only$type_of_violence)
##
## 3
## 21
table(state_only$misalign)
##
## 0 1
## 9 12
table(state_only$scaled)
##
## 1 2 4
## 6 6 9
table(state_only$central_party)
##
## INC NDA NF UPA
## 1 9 2 9
table(state_only$state_party)
##
## AITC BJP CPI(M) INC JMM NPF SAD TDP VACANT YSRCP
## 2 2 4 7 1 1 1 1 1 1
table(state_only$side_a)
##
## Government of India
## 21
table(state_only$side_b)
##
## Civilians
## 21
table(state_only$State)
##
## Andhra Pradesh Assam Bihar Jharkhand Manipur
## 2 3 1 2 1
## Meghalaya Nagaland Punjab Tripura West Bengal
## 1 2 1 4 4
table(state_only$pr)
##
## 0 1
## 20 1
##misalignment and pr occurences
contingency_table <- state_only%>%
count(pr, misalign) %>%
group_by(misalign) %>%
mutate(proportion = n / sum(n)) %>%
ungroup()
# Display the table
print(contingency_table)
## # A tibble: 3 × 4
## pr misalign n proportion
## <dbl> <dbl> <int> <dbl>
## 1 0 0 8 0.889
## 2 0 1 12 1
## 3 1 0 1 0.111
ggplot(contingency_table, aes(x = factor(misalign), y = proportion, fill = factor(pr))) +
geom_bar(stat = "identity", position = position_dodge()) +
labs(title = "Proportion of PR by Alignment",
x = "Alignment",
y = "Proportion",
fill = "PR") +
geom_text(aes(label = scales::percent(proportion), y = proportion + 0.02),
position = position_dodge(width = 0.9),
vjust = 0,
color = "black") +
theme_minimal()
##misalignemnt and afspa
contingency_table_afspa <- state_only%>%
count(afspa, misalign) %>%
group_by(misalign) %>%
mutate(proportion = n / sum(n)) %>%
ungroup()
# Display the table
print(contingency_table_afspa)
## # A tibble: 4 × 4
## afspa misalign n proportion
## <int> <dbl> <int> <dbl>
## 1 0 0 3 0.333
## 2 0 1 11 0.917
## 3 1 0 6 0.667
## 4 1 1 1 0.0833
ggplot(contingency_table_afspa, aes(x = factor(misalign), y = proportion, fill = factor(afspa))) +
geom_bar(stat = "identity", position = position_dodge()) +
labs(title = "Proportion of PR by Alignment",
x = "misalignment",
y = "Proportion",
fill = "afspa") +
geom_text(aes(label = scales::percent(proportion), y = proportion + 0.02),
position = position_dodge(width = 0.9),
vjust = 0,
color = "black") +
theme_minimal()
##summary stats
get_mode <- function(x) {
ux <- unique(x)
ux[which.max(tabulate(match(x, ux)))]
}
mis_stats <- state_only %>%
summarise(
Mean = mean(misalign, na.rm = TRUE),
Median = median(misalign, na.rm = TRUE),
Mode = get_mode(misalign),
Standard_Deviation = sd(misalign, na.rm = TRUE)
)
print(mis_stats)
## Mean Median Mode Standard_Deviation
## 1 0.5714286 1 1 0.5070926
scaled_stats <- state_only %>%
summarise(
Mean = mean(scaled, na.rm = TRUE),
Median = median(scaled, na.rm = TRUE),
Mode = get_mode(scaled),
Standard_Deviation = sd(scaled, na.rm = TRUE)
)
print(scaled_stats)
## Mean Median Mode Standard_Deviation
## 1 2.571429 2 4 1.325573
pr_stats <- state_only %>%
summarise(
Mean = mean(pr, na.rm = TRUE),
Median = median(pr, na.rm = TRUE),
Mode = get_mode(pr),
Standard_Deviation = sd(pr, na.rm = TRUE)
)
print(pr_stats)
## Mean Median Mode Standard_Deviation
## 1 0.04761905 0 0 0.2182179
## sum states of misaligned and fatalities
misal_stats <-state_only %>%
group_by(misalign) %>%
summarise(
Mean_Death = mean(best_est, na.rm = TRUE), # Calculate mean, remove NA values if any
SD_Death = sd(best_est, na.rm = TRUE) # Calculate standard deviation
)
print(misal_stats)
## # A tibble: 2 × 3
## misalign Mean_Death SD_Death
## <dbl> <dbl> <dbl>
## 1 0 1.89 2.52
## 2 1 1.5 1.73
summary_afspa <-state_only %>%
group_by(afspa) %>%
summarise(
Mean_Death = mean(best_est, na.rm = TRUE), # Calculate mean, remove NA values if any
SD_Death = sd(best_est, na.rm = TRUE) # Calculate standard deviation
)
print(summary_afspa)
## # A tibble: 2 × 3
## afspa Mean_Death SD_Death
## <int> <dbl> <dbl>
## 1 0 1.93 2.37
## 2 1 1.14 1.21
summary_state <-state_only %>%
group_by(State) %>%
summarise(
Mean_Death = mean(best_est, na.rm = TRUE), # Calculate mean, remove NA values if any
SD_Death = sd(best_est, na.rm = TRUE) # Calculate standard deviation
)
print(summary_state)
## # A tibble: 10 × 3
## State Mean_Death SD_Death
## <chr> <dbl> <dbl>
## 1 Andhra Pradesh 1 0
## 2 Assam 1 1
## 3 Bihar 0 NA
## 4 Jharkhand 2 1.41
## 5 Manipur 2 NA
## 6 Meghalaya 0 NA
## 7 Nagaland 5.5 3.54
## 8 Punjab 1 NA
## 9 Tripura 1.75 2.87
## 10 West Bengal 1.25 1.26
## Difference in medians test
ww_test_result <- wilcox.test(best_est ~ misalign, data = state_only, exact = FALSE)
print(ww_test_result)
##
## Wilcoxon rank sum test with continuity correction
##
## data: best_est by misalign
## W = 56.5, p-value = 0.8822
## alternative hypothesis: true location shift is not equal to 0
prs_test_result <- wilcox.test(best_est ~ pr, data = state_only, exact = FALSE)
print(prs_test_result)
##
## Wilcoxon rank sum test with continuity correction
##
## data: best_est by pr
## W = 17.5, p-value = 0.228
## alternative hypothesis: true location shift is not equal to 0
afspa_test_result <- wilcox.test(best_est ~ afspa, data = state_only, exact = FALSE)
print(afspa_test_result)
##
## Wilcoxon rank sum test with continuity correction
##
## data: best_est by afspa
## W = 56, p-value = 0.6131
## alternative hypothesis: true location shift is not equal to 0
## difference in medians removing high outliers
filtered_state <- state_only %>%
filter(best_est <= quantile(best_est, 0.95)) # Remove high outliers, keeping up to the 95th percentile
ggplot(filtered_state, aes(x = factor(misalign), y = best_est, fill = factor(misalign))) +
geom_boxplot(outlier.shape = NA) + # Boxplot to show medians and distributions
labs(title = "Fatality Rates by Alignment",
x = "Alignment",
y = "Fatality Rates") +
theme_minimal() +
geom_jitter(width = 0.2, alpha = 0.5, color = "black")
nation<- lm(state_only$misalign ~ state_only$best_est + state_only$State + state_only$pr + state_only$afspa)
summary(nation)
##
## Call:
## lm(formula = state_only$misalign ~ state_only$best_est + state_only$State +
## state_only$pr + state_only$afspa)
##
## Residuals:
## Min 1Q Median 3Q Max
## -0.46830 -0.02536 0.00000 0.02536 0.46830
##
## Coefficients: (1 not defined because of singularities)
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 9.683e-01 1.663e-01 5.822 0.000253 ***
## state_only$best_est 3.170e-02 4.061e-02 0.780 0.455149
## state_only$StateAssam -1.108e+00 3.226e-01 -3.434 0.007459 **
## state_only$StateBihar -9.683e-01 2.823e-01 -3.430 0.007509 **
## state_only$StateJharkhand -5.317e-01 2.317e-01 -2.295 0.047401 *
## state_only$StateManipur -1.139e+00 3.894e-01 -2.926 0.016854 *
## state_only$StateMeghalaya -1.076e+00 3.594e-01 -2.994 0.015103 *
## state_only$StateNagaland -1.197e+00 3.547e-01 -3.373 0.008218 **
## state_only$StatePunjab -6.961e-16 2.794e-01 0.000 1.000000
## state_only$StateTripura -5.071e-02 2.141e-01 -0.237 0.818084
## state_only$StateWest Bengal -7.924e-03 1.978e-01 -0.040 0.968921
## state_only$pr NA NA NA NA
## state_only$afspa 1.078e-01 2.464e-01 0.437 0.672106
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 0.2281 on 9 degrees of freedom
## Multiple R-squared: 0.9089, Adjusted R-squared: 0.7976
## F-statistic: 8.167 on 11 and 9 DF, p-value: 0.001943
## insignificant results
##################################################################
state_actor<-ds%>%
filter(type_of_violence == 1)
# Perform a Mann-Whitney U test
mw_test_result <- wilcox.test(best_est ~ misalign, data = state_actor, exact = FALSE)
print(mw_test_result)
##
## Wilcoxon rank sum test with continuity correction
##
## data: best_est by misalign
## W = 980087, p-value = 0.005363
## alternative hypothesis: true location shift is not equal to 0
afspa_test_result <- wilcox.test(best_est ~ afspa, data = state_actor, exact = FALSE)
print(afspa_test_result)
##
## Wilcoxon rank sum test with continuity correction
##
## data: best_est by afspa
## W = 1007784, p-value = 0.05668
## alternative hypothesis: true location shift is not equal to 0
pr_test_result <- wilcox.test(best_est ~ pr, data = state_actor, exact = FALSE)
print(pr_test_result)
##
## Wilcoxon rank sum test with continuity correction
##
## data: best_est by pr
## W = 325454, p-value = 1.895e-07
## alternative hypothesis: true location shift is not equal to 0
mis_stats_act <- state_actor %>%
summarise(
Mean = mean(misalign, na.rm = TRUE),
Median = median(misalign, na.rm = TRUE),
Mode = get_mode(misalign),
Standard_Deviation = sd(misalign, na.rm = TRUE)
)
print(mis_stats_act)
## Mean Median Mode Standard_Deviation
## 1 0.3928209 0 0 0.4884603
scaled_stats_act <- state_actor %>%
summarise(
Mean = mean(scaled, na.rm = TRUE),
Median = median(scaled, na.rm = TRUE),
Mode = get_mode(scaled),
Standard_Deviation = sd(scaled, na.rm = TRUE)
)
print(scaled_stats_act)
## Mean Median Mode Standard_Deviation
## 1 2.870301 4 4 1.321508
pr_stats_act <- state_actor %>%
summarise(
Mean = mean(pr, na.rm = TRUE),
Median = median(pr, na.rm = TRUE),
Mode = get_mode(pr),
Standard_Deviation = sd(pr, na.rm = TRUE)
)
print(pr_stats_act)
## Mean Median Mode Standard_Deviation
## 1 0.1005757 0 0 0.300817
## sum states of misaligned and fatalities
misal_stats_only <-state_actor %>%
group_by(misalign) %>%
summarise(
Mean_Death = mean(best_est, na.rm = TRUE), # Calculate mean, remove NA values if any
SD_Death = sd(best_est, na.rm = TRUE) # Calculate standard deviation
)
print(misal_stats_only)
## # A tibble: 2 × 3
## misalign Mean_Death SD_Death
## <dbl> <dbl> <dbl>
## 1 0 5.59 72.3
## 2 1 2.59 2.81
summary_afspa_act <-state_actor %>%
group_by(afspa) %>%
summarise(
Mean_Death = mean(best_est, na.rm = TRUE), # Calculate mean, remove NA values if any
SD_Death = sd(best_est, na.rm = TRUE) # Calculate standard deviation
)
print(summary_afspa_act)
## # A tibble: 2 × 3
## afspa Mean_Death SD_Death
## <int> <dbl> <dbl>
## 1 0 2.82 3.38
## 2 1 5.20 69.0
summary_state_act <-state_actor %>%
group_by(State, misalign) %>%
summarise(
Mean_Death = mean(best_est, na.rm = TRUE), # Calculate mean, remove NA values if any
SD_Death = sd(best_est, na.rm = TRUE) # Calculate standard deviation
)
## `summarise()` has grouped output by 'State'. You can override using the
## `.groups` argument.
view(summary_state_act)
####
filtered_state_actor <- state_actor %>%
filter(best_est <= quantile(best_est, 0.95)) # Remove high outliers, keeping up to the 95th percentile
ggplot(filtered_state_actor, aes(x = factor(misalign), y = best_est, fill = factor(misalign))) +
geom_boxplot(outlier.shape = NA) + # Boxplot to show medians and distributions
labs(title = "Fatality Rates by Alignment",
x = "Alignment",
y = "Fatality Rates") +
theme_minimal() +
geom_jitter(width = 0.2, alpha = 0.5, color = "black")
table(state_only$type_of_violence)
##
## 3
## 21
table(state_only$align)
## < table of extent 0 >
table(state_only$scaled)
##
## 1 2 4
## 6 6 9
table(state_only$central_party)
##
## INC NDA NF UPA
## 1 9 2 9
table(state_only$state_party)
##
## AITC BJP CPI(M) INC JMM NPF SAD TDP VACANT YSRCP
## 2 2 4 7 1 1 1 1 1 1
table(state_only$side_a)
##
## Government of India
## 21
table(state_only$side_b)
##
## Civilians
## 21
table(state_only$State)
##
## Andhra Pradesh Assam Bihar Jharkhand Manipur
## 2 3 1 2 1
## Meghalaya Nagaland Punjab Tripura West Bengal
## 1 2 1 4 4
table(state_only$pr)
##
## 0 1
## 20 1
contingency_table_actor <- state_actor%>%
count(pr, misalign) %>%
group_by(misalign) %>%
mutate(proportion = n / sum(n)) %>%
ungroup()
# Display the table
print(contingency_table_actor)
## # A tibble: 4 × 4
## pr misalign n proportion
## <dbl> <dbl> <int> <dbl>
## 1 0 0 1533 0.855
## 2 0 1 1123 0.968
## 3 1 0 260 0.145
## 4 1 1 37 0.0319
ggplot(contingency_table_actor, aes(x = factor(misalign), y = proportion, fill = factor(pr))) +
geom_bar(stat = "identity", position = position_dodge()) +
labs(title = "Proportion of PR by Alignment",
x = "misalignment",
y = "Proportion",
fill = "PR") +
geom_text(aes(label = scales::percent(proportion), y = proportion + 0.02),
position = position_dodge(width = 0.9),
vjust = 0,
color = "black") +
theme_minimal()
#kruskal wallis test
kruskal.test(best_est ~ misalign, data = state_actor)
##
## Kruskal-Wallis rank sum test
##
## data: best_est by misalign
## Kruskal-Wallis chi-squared = 7.7531, df = 1, p-value = 0.005362
library(dunn.test)
dunn_result <- dunn.test(state_actor$best_est, state_actor$misalign, method = "bonferroni")
## Kruskal-Wallis rank sum test
##
## data: x and group
## Kruskal-Wallis chi-squared = 7.7531, df = 1, p-value = 0.01
##
##
## Comparison of x by group
## (Bonferroni)
## Col Mean-|
## Row Mean | 0
## ---------+-----------
## 1 | -2.784430
## | 0.0027*
##
## alpha = 0.05
## Reject Ho if p <= alpha/2
print(dunn_result)
## $chi2
## [1] 7.753052
##
## $Z
## [1] -2.78443
##
## $P
## [1] 0.002681093
##
## $P.adjusted
## [1] 0.002681093
##
## $comparisons
## [1] "0 - 1"
Q1 <- quantile(state_actor$best_est, 0.25)
Q3 <- quantile(state_actor$best_est, 0.75)
IQR <- IQR(state_actor$best_est)
ggplot(state_actor, aes(x = factor(misalign), y = best_est, fill = factor(misalign))) +
geom_boxplot() +
coord_cartesian(ylim = c(Q1 - 1.5 * IQR, Q3 + 1.5 * IQR)) + # Adjust axis limits to IQR range
labs(title = "Comparison of Best Estimates by Misalignment Group",
x = "Misalignment Group",
y = "Best Estimates") +
theme_minimal()
ggplot(state_actor, aes(x = factor(pr), y = best_est, fill = factor(pr))) +
geom_boxplot() +
coord_cartesian(ylim = c(Q1 - 1.5 * IQR, Q3 + 1.5 * IQR)) + # Adjust axis limits to IQR range
labs(title = "Comparison of Best Estimates by PR",
x = "PR",
y = "Best Estimates") +
theme_minimal()
ggplot(state_actor, aes(x = factor(afspa), y = best_est, fill = factor(afspa))) +
geom_boxplot() +
coord_cartesian(ylim = c(Q1 - 1.5 * IQR, Q3 + 1.5 * IQR)) + # Adjust axis limits to IQR range
labs(title = "Comparison of Best Estimates by AFSPA",
x = "AFSPA",
y = "Best Estimates") +
theme_minimal()
# Assuming state_actor is your dataframe, and it has the columns year, misalign, and best_est
sum_year <- state_actor %>%
group_by(Year, misalign) %>%
summarise(
median_best_est = median(best_est, na.rm = TRUE),
lower_quartile = quantile(best_est, 0.25, na.rm = TRUE),
upper_quartile = quantile(best_est, 0.75, na.rm = TRUE)
) %>%
ungroup()
## `summarise()` has grouped output by 'Year'. You can override using the
## `.groups` argument.
ggplot(sum_year, aes(x = factor(misalign), y = median_best_est, fill = factor(Year))) +
geom_bar(stat = "identity", position = "dodge") +
geom_errorbar(aes(ymin = lower_quartile, ymax = upper_quartile), width = 0.2, position = position_dodge(0.9)) +
labs(title = "Median Fatality Rates by Misalignment and Year",
x = "Misalignment",
y = "Median Fatality Rates",
fill = "Year") +
theme_minimal()
sum_state <- state_actor %>%
group_by(State, misalign) %>%
summarise(
median_best_est = median(best_est, na.rm = TRUE),
lower_quartile = quantile(best_est, 0.25, na.rm = TRUE),
upper_quartile = quantile(best_est, 0.75, na.rm = TRUE)
) %>%
ungroup()
## `summarise()` has grouped output by 'State'. You can override using the
## `.groups` argument.
ggplot(sum_state, aes(x = factor(misalign), y = median_best_est, fill = factor(State))) +
geom_bar(stat = "identity", position = "dodge") +
geom_errorbar(aes(ymin = lower_quartile, ymax = upper_quartile), width = 0.2, position = position_dodge(0.9)) +
labs(title = "Median Fatality Rates by Misalignment and State",
x = "Misalignment",
y = "Median Fatality Rates",
fill = "State") +
theme_minimal()
ggplot(sum_year, aes(x = Year, y = median_best_est, group = misalign, color = factor(misalign))) +
geom_line() +
labs(title = "Median Fatality Rates by Year and Misalignment",
x = "Year",
y = "Median Fatality Rates",
color = "Misalignment") +
theme_minimal()
bjp<- state_actor%>%
filter(central_party%in% c('BJP','NDA'))
summary(bjp)
## id relid Year active_year
## Length:1032 Length:1032 Length:1032 Length:1032
## Class :character Class :character Class :character Class :character
## Mode :character Mode :character Mode :character Mode :character
##
##
##
##
## code_status type_of_violence conflict_dset_id conflict_new_id
## Length:1032 Length:1032 Length:1032 Length:1032
## Class :character Class :character Class :character Class :character
## Mode :character Mode :character Mode :character Mode :character
##
##
##
##
## conflict_name dyad_dset_id dyad_new_id dyad_name
## Length:1032 Length:1032 Length:1032 Length:1032
## Class :character Class :character Class :character Class :character
## Mode :character Mode :character Mode :character Mode :character
##
##
##
##
## side_a_dset_id side_a_new_id side_a side_b_dset_id
## Length:1032 Length:1032 Length:1032 Length:1032
## Class :character Class :character Class :character Class :character
## Mode :character Mode :character Mode :character Mode :character
##
##
##
##
## side_b_new_id side_b number_of_sources source_article
## Length:1032 Length:1032 Min. :-1.0000 Length:1032
## Class :character Class :character 1st Qu.:-1.0000 Class :character
## Mode :character Mode :character Median :-1.0000 Mode :character
## Mean : 0.1541
## 3rd Qu.: 1.0000
## Max. : 7.0000
##
## source_office source_date source_headline source_original
## Length:1032 Length:1032 Length:1032 Length:1032
## Class :character Class :character Class :character Class :character
## Mode :character Mode :character Mode :character Mode :character
##
##
##
##
## where_prec where_coordinates where_description State
## Min. :1.000 Length:1032 Length:1032 Length:1032
## 1st Qu.:2.000 Class :character Class :character Class :character
## Median :3.000 Mode :character Mode :character Mode :character
## Mean :2.461
## 3rd Qu.:3.000
## Max. :5.000
##
## adm_2 latitude longitude geom_wkt
## Length:1032 Length:1032 Length:1032 Length:1032
## Class :character Class :character Class :character Class :character
## Mode :character Mode :character Mode :character Mode :character
##
##
##
##
## priogrid_gid country country_id region
## Length:1032 Length:1032 Length:1032 Length:1032
## Class :character Class :character Class :character Class :character
## Mode :character Mode :character Mode :character Mode :character
##
##
##
##
## event_clarity date_prec date_start date_end
## Length:1032 Min. :1.00 Length:1032 Length:1032
## Class :character 1st Qu.:1.00 Class :character Class :character
## Mode :character Median :1.00 Mode :character Mode :character
## Mean :1.09
## 3rd Qu.:1.00
## Max. :5.00
##
## deaths_a deaths_b deaths_civilians deaths_unknown
## Min. : 0.0000 Min. : 0.00 Min. : 0.0000 Min. : 0.00000
## 1st Qu.: 0.0000 1st Qu.: 0.00 1st Qu.: 0.0000 1st Qu.: 0.00000
## Median : 0.0000 Median : 1.00 Median : 0.0000 Median : 0.00000
## Mean : 0.9331 Mean : 1.39 Mean : 0.1764 Mean : 0.06977
## 3rd Qu.: 1.0000 3rd Qu.: 2.00 3rd Qu.: 0.0000 3rd Qu.: 0.00000
## Max. :26.0000 Max. :14.00 Max. :14.0000 Max. :18.00000
##
## best_est high_est low_est civilian_fatalities
## Min. : 0.000 Min. : 1.000 Min. : 0.000 Min. : 0.0
## 1st Qu.: 1.000 1st Qu.: 1.000 1st Qu.: 1.000 1st Qu.: 15.0
## Median : 2.000 Median : 2.000 Median : 2.000 Median : 35.0
## Mean : 2.569 Mean : 2.763 Mean : 2.512 Mean : 213.4
## 3rd Qu.: 3.000 3rd Qu.: 3.000 3rd Qu.: 3.000 3rd Qu.: 224.0
## Max. :26.000 Max. :26.000 Max. :26.000 Max. :1024.0
##
## sf_fatalities insurgent_fatalities afspa pr
## Min. : 0.0 Min. : 0.0 Min. :0.000 Min. :0.00000
## 1st Qu.: 8.0 1st Qu.: 23.0 1st Qu.:0.000 1st Qu.:0.00000
## Median : 25.0 Median : 68.0 Median :1.000 Median :0.00000
## Mean :134.7 Mean : 486.8 Mean :0.626 Mean :0.01163
## 3rd Qu.:248.0 3rd Qu.: 262.0 3rd Qu.:1.000 3rd Qu.:0.00000
## Max. :628.0 Max. :2345.0 Max. :1.000 Max. :1.00000
##
## duration reasons state_party central_party
## Min. : 26.0 Length:1032 Length:1032 Length:1032
## 1st Qu.: 81.5 Class :character Class :character Class :character
## Median :156.0 Mode :character Mode :character Mode :character
## Mean :137.5
## 3rd Qu.:212.0
## Max. :212.0
## NA's :1020
## misalignment scaled X_merge state
## Min. :0.0000 Min. :1.000 Length:1032 Length:1032
## 1st Qu.:1.0000 1st Qu.:1.000 Class :character Class :character
## Median :1.0000 Median :1.000 Mode :character Mode :character
## Mean :0.8091 Mean :1.755
## 3rd Qu.:1.0000 3rd Qu.:2.000
## Max. :1.0000 Max. :4.000
##
## misalign
## Min. :0.000
## 1st Qu.:1.000
## Median :1.000
## Mean :0.813
## 3rd Qu.:1.000
## Max. :1.000
##
table(bjp$misalign)
##
## 0 1
## 193 839
table(bjp$pr)
##
## 0 1
## 1020 12
table(bjp$afspa)
##
## 0 1
## 386 646
table(bjp$state_party)
##
## AGP AITC BJD BJP CPI(M) INC IND JDU JMM MSCP NDPP
## 108 1 69 113 88 354 3 40 44 30 7
## NPF NPP RJD SAD SP TDP UDP VACANT YSRCP
## 13 1 32 5 6 105 1 6 6
bjp_reg<- lm(best_est ~ misalign + State, data = bjp)
summary(bjp_reg)
##
## Call:
## lm(formula = best_est ~ misalign + State, data = bjp)
##
## Residuals:
## Min 1Q Median 3Q Max
## -3.4941 -1.4679 -0.6959 0.5671 22.7985
##
## Coefficients:
## Estimate Std. Error t value Pr(>|t|)
## (Intercept) 2.4679 0.3553 6.946 6.7e-12 ***
## misalign 0.2280 0.2799 0.815 0.41546
## StateArunachal Pradesh -1.3700 0.6952 -1.971 0.04903 *
## StateAssam -0.4943 0.2838 -1.742 0.08187 .
## StateBihar 0.5974 0.4256 1.404 0.16068
## StateJharkhand 0.7337 0.3865 1.898 0.05794 .
## StateManipur 0.7982 0.3782 2.111 0.03506 *
## StateMeghalaya -1.2619 0.4878 -2.587 0.00982 **
## StateNagaland 0.3296 0.6708 0.491 0.62330
## StateOdisha -0.7829 0.4116 -1.902 0.05744 .
## StatePunjab 0.5321 1.2667 0.420 0.67452
## StateTripura 0.4013 0.3833 1.047 0.29540
## StateWest Bengal -0.1959 1.3821 -0.142 0.88731
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## Residual standard error: 2.719 on 1019 degrees of freedom
## Multiple R-squared: 0.04641, Adjusted R-squared: 0.03518
## F-statistic: 4.133 on 12 and 1019 DF, p-value: 2.395e-06
sum_bjp <- bjp %>%
group_by(State, misalign) %>%
summarise(
median_best_est = median(best_est, na.rm = TRUE),
lower_quartile = quantile(best_est, 0.25, na.rm = TRUE),
upper_quartile = quantile(best_est, 0.75, na.rm = TRUE)
) %>%
ungroup()
## `summarise()` has grouped output by 'State'. You can override using the
## `.groups` argument.
ggplot(sum_bjp, aes(x = factor(misalign), y = median_best_est, fill = factor(State))) +
geom_bar(stat = "identity", position = "dodge") +
geom_errorbar(aes(ymin = lower_quartile, ymax = upper_quartile), width = 0.2, position = position_dodge(0.9)) +
labs(title = "Median Fatality Rates by Misalignment and State",
x = "Misalignment",
y = "Median Fatality Rates",
fill = "State") +
theme_minimal()
bjp_year <- bjp %>%
group_by(Year, misalign) %>%
summarise(
median_best_est = median(best_est, na.rm = TRUE),
lower_quartile = quantile(best_est, 0.25, na.rm = TRUE),
upper_quartile = quantile(best_est, 0.75, na.rm = TRUE)
) %>%
ungroup()
## `summarise()` has grouped output by 'Year'. You can override using the
## `.groups` argument.
ggplot(bjp_year, aes(x = factor(misalign), y = median_best_est, fill = factor(Year))) +
geom_bar(stat = "identity", position = "dodge") +
geom_errorbar(aes(ymin = lower_quartile, ymax = upper_quartile), width = 0.2, position = position_dodge(0.9)) +
labs(title = "Median Fatality Rates by Misalignment and Year",
x = "Misalignment",
y = "Median Fatality Rates",
fill = "Year") +
theme_minimal()
ggplot(sum_year, aes(x = Year, y = median_best_est, group = misalign, color = factor(misalign))) +
geom_line() +
labs(title = "Median Fatality Rates by Year and Misalignment",
x = "Year",
y = "Median Fatality Rates",
color = "Misalignment") +
theme_minimal()
ggplot(bjp_year, aes(x = Year, y = median_best_est, group = misalign, color = factor(misalign))) +
geom_line() +
labs(title = "Median Fatality Rates by Year and Misalignment",
x = "Year",
y = "Median Fatality Rates",
color = "Misalignment") +
theme_minimal()
