Final Project 607

Introduction

In recent decades, society has been the victim of one of the worst scourges: terrorism, which is an expressive fact of violence that can be seen throughout history with its most varied forms of expression and cruelty. Numerous terrorist incidents within the hemisphere during the last years have shown that terrorism is an extreme and constant danger for local and worldwide peace. This phenomenon is one of the most difficult forms of violence to contain because its scope extends beyond the regions of conflict. It is a phenomenon characterized by its indiscriminate violence which involves victims who have nothing to do with the conflict that caused the terrorist act. Its unpredictability acts by surprise creating uncertainty and instilling terror on people. The consequences of those activities for the safety of human rights and democracy are extraordinarily critical and require immediate and rigorous consideration by means of the global community, such as the Organization of American States. As this commission has reiterated, international law obliges the member states to adopt the essential measures to prevent terrorism and other forms of violence and to assure the security in their residents.

Abstract

The Global Terrorism Database (GTD) was developed to be a comprehensive, methodologically robust set of longitudinal data on incidents of domestic and international terrorism. Its primary purpose is to enable researchers and analysts to increase understanding of the phenomenon of terrorism. The GTD is specifically designed to be amenable to the latest quantitative analytic techniques used in the social and computational sciences.

The GTD was designed to gather a wide variety of etiological and situational variables pertaining to each terrorist incident. Depending on availability of information, the database records up to 120 separate attributes of each incident, including approximately 75 coded variables that can be used for statistical analysis.

Resources

https://www.start.umd.edu/gtd/

Research Question

For this project, the main goal is to summarise the data pertaining to terrorist attacks in order to explain the proportional effect of terrorism in the world as well as the major trends related to their activities.

Data Preparation

This sections defines basic cleaning operations. Other cleaning tasks are going to be performed as needed accross the sections.

Loading the dataset.

terror_dt <- read.csv("globalterrorismdb_0718dist.csv", stringsAsFactors = TRUE)

Loading world map dataset for ploting later.

world_dt <- map_data("world") %>% fortify()
map_locations <- terror_dt[(!is.na(terror_dt$latitude) & !is.na(terror_dt$longitude)),]

Filling Values for Columns representing the number of victims and injured where Na values suggest that no victims where recorded, therefore, we assume that there were no victims or injured

terror_dt$nkill %<>% replace_na(0) 
terror_dt$nwound %<>% replace_na(0) 

Exploratory Analysis

In this section, we are going to observe what are the facts and trents that are stated in the Global Terrorism dataset.

World Map of Terrorist Attacks

This world map illustratres all the places around the world were incidents classified as terrorist attack due to a especific criteria.

locations <- geom_point(data = map_locations,aes(x = longitude,y = latitude,color=region_txt),size=1)


world_plot <- ggplot()+geom_polygon(
  data = world_dt,
  aes(x = long,
      y = lat,
    group=group,
    map_id=region),
  fill="#4B535D",
  colour="white",
  size=0.1)+xlim(-170,190)+ylim(-60,90)+ggtitle("World Map of terrorist Incidents from 1970 to 2017")+xlab("Longitude")+ylab("Latitude")+theme(legend.title = element_blank(),legend.position = "bottom",panel.background = element_rect(fill="#CCD3D3"),panel.grid.major = element_blank(),panel.grid.minor = element_blank(),legend.text = element_text(size = 13))

world_plot+locations

As a quick reference, each region in the world map groups the following countries:

North America: includes countris such as Canada, Mexico, United States.

Central America & Caribbean: includes countries such as Antigua and Barbuda, Bahamas, Barbados, Belize, Cayman Islands, Costa Rica, Cuba, Dominica, Dominican Republic, El Salvador, Grenada, Guadeloupe, Guatemala, Haiti, Honduras, Jamaica, Martinique, Nicaragua, Panama, St. Kitts and Nevis, St. Lucia,Trinidad and Tobago

South America: include countries such as Argentina, Bolivia, Brazil, Chile, Colombia, Ecuador, Falkland Islands, French Guiana, Guyana, Paraguay, Peru, Suriname, Uruguay, Venezuela

East Asia: includes China, Hong Kong, Japan, Macau, North Korea, South Korea, Taiwan

Southeast Asia: includes Brunei, Cambodia, East Timor, Indonesia, Laos, Malaysia, Myanmar, Philippines, Singapore, South Vietnam, Thailand, Vietnam

South Asia: includes Afghanistan, Bangladesh, Bhutan, India, Maldives, Mauritius, Nepal, Pakistan, Sri Lanka

Central Asia: includes Armenia, Azerbaijan, Georgia, Kazakhstan, Kyrgyzstan, Tajikistan, Turkmenistan, Uzbekistan

Western Europe: includes Andorra, Austria, Belgium, Cyprus, Denmark, Finland, France, Germany, Gibraltar, Greece, Iceland, Ireland, Italy, Luxembourg, Malta, Netherlands, Norway, Portugal, Spain, Sweden, Switzerland, United Kingdom, Vatican City, West Germany (FRG)

Eastern Europe: includes Albania, Belarus, Bosnia-Herzegovina, Bulgaria, Croatia, Czech Republic, Czechoslovakia, East Germany (GDR), Estonia, Hungary, Kosovo, Latvia, Lithuania, Macedonia, Moldova, Montenegro, Poland, Romania, Russia, Serbia, Serbia- Montenegro, Slovak Republic, Slovenia, Soviet Union, Ukraine, Yugoslavia

Middle East & North Africa: includes Algeria, Bahrain, Egypt, Iran, Iraq, Israel, Jordan, Kuwait, Lebanon, Libya, Morocco, North Yemen, Qatar, Saudi Arabia, South Yemen, Syria, Tunisia, Turkey, United Arab Emirates, West Bank and Gaza Strip, Western Sahara, Yemen

Sub-Saharan Africa: includes Angola, Benin, Botswana, Burkina Faso, Burundi, Cameroon, Central African Republic, Chad, Comoros, Democratic Republic of the Congo, Djibouti, Equatorial Guinea, Eritrea, Ethiopia, Gabon, Gambia, Ghana, Guinea, Guinea-Bissau, Ivory Coast, Kenya, Lesotho, Liberia, Madagascar, Malawi, Mali, Mauritania, Mozambique, Namibia, Niger, Nigeria, People’s Republic of the Congo, Republic of the Congo, Rhodesia, Rwanda, Senegal, Seychelles, Sierra Leone, Somalia, South Africa, South Sudan, Sudan, Swaziland, Tanzania, Togo, Uganda, Zaire, Zambia, Zimbabwe

Australasia & Oceania: includes Australia, Fiji, French Polynesia, New Caledonia, New Hebrides, New Zealand, Papua, New Guinea, Solomon Islands, Vanuatu, Wallis and Futuna

Incidents by country

As shown in the graph, the higer number of incidents in the world belong to Iraq from Middle East & North Africa, followed by countries such as Pakistan, Afganistan, the India from South Asia and Colombia from South America.

terror_no <- terror_dt %>% group_by(country_txt) %>% tally(name = "attacks",sort = TRUE)

terror_no_plt <- terror_no %>% top_n(20) %>% ggplot(aes(x = fct_reorder(country_txt,attacks),y = attacks,fill=country_txt))+geom_bar(stat = "identity")+coord_flip()+theme(legend.position = "none")+ggtitle("Top 20 of countries with more incidents")+xlab("countries")+geom_text(aes(label=attacks), hjust=1)


terror_no_plt

The table below summarises the number of attacks by country.

datatable(terror_no,colnames = c("Rank","Country","Number of Attacks"),caption = "Table 1: Number of incidents for each country from 1970 to 2017")

Cities in the World

The following graph represents the cities in the world with the higer number of incidents. Baghdad and Mosul in Iraq, Karachi in Pakistan, Lima in Peru, and Belfast in the United Kingdom are the cities in the world witht the top five places of terrorist incidents.

city_no <- terror_dt %>% select(city,region_txt,country_txt) %>% group_by(city,region_txt,country_txt) %>% tally(name = "count",sort = TRUE) %>% ungroup()


city_no[city_no$city!="Unknown",] %>% top_n(20) %>%  ggplot(aes(x = fct_reorder(city,count),y = count,fill=region_txt))+geom_bar(stat = "identity")+coord_flip()+theme(legend.title = element_blank(),legend.position = "bottom")+ggtitle("Top 20 of Cities in the World with more Incidents")+xlab("countries")

The table below summarises the number of incidents for a particular city in the world organized by the number of attacks.

city_incidents <- city_no[city_no$city != "Unknown",]

datatable(city_incidents,colnames = c("Rank","Country","Region","Number of Attacks"),caption = "Table 2: Cities in the world with more incidents")

Incidents by year

As show in the graph, the number of terrorist incidents started to increase dramatically reaching a high number of incidents between 2003 and 2015 after a trough before 1997.

#Number of attacks for earch region during since 1970 to 2017
attack_tml <- terror_dt %>% group_by(region_txt,iyear) %>% tally(name = "incident_no")

#######

atk_tml_plt <- attack_tml %>% ggplot(aes(x = iyear,y = incident_no,group=region_txt,color=region_txt))+geom_line()+ggtitle("Number of Incidents by year")+ylab("Number of Incidents")+xlab("Year")+labs(colour="Region")+scale_x_discrete(breaks=seq(1970,2017,by=5),limits=1970:2017)+theme(legend.position = "bottom",legend.text = element_text(size = 12))

atk_tml_plt

atk_tml_plt+facet_wrap(region_txt~.,scales = "free_y",ncol = 2)+theme(legend.position = "none")

• Central Asia had experience a maximum of 80 incidents in 1992. after that, the number of incidents have been decreasing with some light increases in the number of incidents until 2017.

• Sub-Sahara Africa had a number of incidents that did not passed the threshold of 500 from 1970 to 2010. After that the number of incidents started to increase reaching a number grater than 1500 incidents for the rest of the years.

• North America, from 1970 to 1973 seemed to be years of high terrorist activity, however, it started to decrease intil it reaches a certain level of estability from 1974 to 2017.

• From 1975 to 2000 in Central America & Caribbean, the activity of terrorism was relatively high, however, the activity of incidents decreased to near 0 for the last years.

• South America Experienced high level of terrorist activity until 2000, then the levels of incidents decreased and keep a somewhat normal level of activity.

• Australia and Oceania had an active level of incidents from 1970 to 2017, however, the maximum number of incidents that happened during those years is 30; therefore, it is verylow compared to other regions of the world.

Fatalities by Region

This section includes the number of victims and attackers who died as a direct result of the incident. Where there is evidence of fatalities, but a figure is not reported or it is too vague to be of use, this field is counted as 0. If information is missing regarding the number of victims killed in an attack, but perpetrator fatalities are known, the number of victims will reflect only the number of perpetrators who died as a result of the incident and viceversa.

#Number of kills for earch region during since 1970 to 2017
victim_tml <- terror_dt %>% select(region_txt,nkill,iyear) %>% group_by(region_txt,iyear) %>% summarise(victims=sum(nkill)) %>% ungroup()


#victim_tml$victims[is.na(victim_tml$victims)] <- 0

tml <- victim_tml%>% ggplot(aes(x = as.numeric(iyear),y = victims,group=region_txt))+geom_line(aes(color=region_txt))+theme(axis.text.x = element_text(angle = 50),legend.position = "bottom",legend.text = element_text(size = 10))+scale_x_discrete(breaks=seq(1970,2017,by=5),limits=1970:2017)+xlab("years")+ggtitle("Victims Timeline")

tml

Detailed View

tml+facet_wrap(.~region_txt,ncol = 1, scales = "free_y")+theme(legend.position = "none",panel.spacing = unit(3,"line"))

nkill_summary <- terror_dt %>% group_by(region_txt) %>% select(region_txt,nkill,city)


nkill_summary %<>% summarise(total_kill=sum(nkill)) %>% mutate(proportion=round(total_kill/sum(total_kill)*100,3))

nkill_summary %<>% mutate_if(is.character,as.numeric)
nkill_summary %<>% mutate_if(is.factor,as.character)

nkill_summary %<>% rbind(c("Total",apply(nkill_summary[,2:3],2,sum)))

nkill_summary  %>% 
  kable(col.names = c("Region","Fatalities Number","World Percentaje"),caption = "Table 3: Number of fatalities from 1970 to 2017") %>% 
  kable_styling(position = "center",full_width = F,bootstrap_options = "striped",font_size = 13)

Table 3: Number of fatalities from 1970 to 2017

Region	Fatalities Number	World Percentaje
Australasia & Oceania	150	0.036
Central America & Caribbean	28708	6.97
Central Asia	1000	0.243
East Asia	1152	0.28
Eastern Europe	7415	1.8
Middle East & North Africa	137642	33.419
North America	4916	1.194
South America	28849	7.004
South Asia	101319	24.6
Southeast Asia	15637	3.797
Sub-Saharan Africa	78386	19.032
Western Europe	6694	1.625
Total	411868	100

Number of Injured

This section records the number of confirmed non-fatal injuries to both perpetrators and victims around the world

#Number of people injured for earch region during since 1970 to 2017
injured_tml <- terror_dt %>% select(region_txt,nwound,iyear) %>% group_by(region_txt,iyear) %>% summarise(injured=sum(nwound)) %>% ungroup()

injured_tml %>% ggplot(aes(y = injured,x = iyear,group=region_txt,color=region_txt))+geom_line()+theme(axis.text.x = element_text(angle = 50),legend.position = "bottom",legend.text = element_text(size = 10),legend.title = element_blank())+scale_x_discrete(breaks=seq(1970,2017,by=5),limits=1970:2017)+xlab("years")+ggtitle("Number of Injured People over the Years")

As shown in the graph above, the number of injured due to terrorism from 1990s to 2017 started to reach higer peaks compared to previous years

wound_summary <- terror_dt %>% group_by(region_txt) %>%
  summarise(totalw=sum(nwound)) %>% 
  mutate(proportion=round(totalw/sum(totalw)*100,3))



wound_summary %<>% mutate_if(is.character,as.numeric)
wound_summary %<>% mutate_if(is.factor,as.character)

wound_summary %<>% rbind(c("Total",apply(wound_summary[,2:3],2,sum)))


wound_summary   %>% 
  kable(col.names = c("Region","Injured Number","World Percentaje"),caption = "Table 4: Number of injured from 1970 to 2017 for each region of the world") %>% 
  kable_styling(position = "center",full_width = F,bootstrap_options = "striped",font_size = 13)

Table 4: Number of injured from 1970 to 2017 for each region of the world

Region	Injured Number	World Percentaje
Australasia & Oceania	260	0.05
Central America & Caribbean	8991	1.716
Central Asia	2009	0.383
East Asia	9213	1.759
Eastern Europe	12045	2.299
Middle East & North Africa	214308	40.909
North America	21531	4.11
South America	16704	3.189
South Asia	141360	26.984
Southeast Asia	26259	5.013
Sub-Saharan Africa	52857	10.09
Western Europe	18332	3.499
Total	523869	100.001

Targets of Terrorism

This sections shows the general type of target or victim. When a victim is attacked specifically because of his or her relationship to a particular person, such as a prominent figure, the target type reflects that motive. For example, if a family member of a government official is attacked because of his or her relationship to that individual, the type of target is “government.”

This variable consists of the following 22 categories which are shown in the following graphs which display the occurence of incidents categorized according to the type of target that terrorist wanted to reach by region.

targets <- terror_dt %>% select(targtype1_txt,nkill,region_txt) %>% group_by(targtype1_txt,region_txt) %>% tally(name = "incidents_no",sort = TRUE)


targets %>% ggplot(aes(x = region_txt,y = incidents_no,fill=targtype1_txt))+geom_bar(stat = "identity")+coord_flip()+facet_wrap(.~ targtype1_txt,ncol = 1,scales = "free_x")+theme(legend.position = "none",plot.background = element_rect(fill = "transparent",colour = NA)
)+xlab("")+ylab("")+ggtitle("Incident Type Occurence by Region from 1970 to 2017")

This table summarises the proportion of attacks around the world targeted to an especific type of victim.

total_target <- targets$incidents_no %>% sum()


target_summary <- targets %>% 
  spread(key = region_txt,value = incidents_no, fill = 0) %>% ungroup() 



target_summary[,2:length(target_summary)] <- round((target_summary[,2:length(target_summary)]/total_target)*100,3)

target_summary %<>%  mutate(Total=rowSums(.[2:13]))

target_summary[[1]] <- target_summary[[1]] %>% as.character()


total <- c("Total",apply(target_summary[,2:14], 2,FUN = sum))

target_summary <- target_summary %>% rbind(total)

header <- names(target_summary) %>% unlist()
header[1] <- "Region"
target_summary %>% 
  kable(caption = "Table 5: Proportion of Incidents by Region",col.names = header) %>% 
  kable_styling(position = "left",full_width = F,bootstrap_options = "striped",font_size = 13) %>% scroll_box(width = "100%")

Table 5: Proportion of Incidents by Region

Region	Australasia & Oceania	Central America & Caribbean	Central Asia	East Asia	Eastern Europe	Middle East & North Africa	North America	South America	South Asia	Southeast Asia	Sub-Saharan Africa	Western Europe	Total
Abortion Related	0	0	0	0	0	0	0.142	0.001	0	0	0	0.002	0.145
Airports & Aircraft	0.004	0.041	0.003	0.032	0.017	0.152	0.031	0.08	0.096	0.028	0.073	0.182	0.739
Business	0.025	0.656	0.019	0.056	0.239	2.289	0.498	1.923	1.719	0.951	0.685	2.317	11.377
Educational Institution	0.004	0.113	0.004	0.008	0.032	0.451	0.098	0.195	0.98	0.225	0.171	0.098	2.379
Food or Water Supply	0.001	0.009	0.001	0.001	0.005	0.03	0.004	0.044	0.035	0.013	0.025	0.008	0.176
Government (Diplomatic)	0.017	0.121	0.018	0.019	0.075	0.44	0.101	0.215	0.166	0.068	0.356	0.372	1.968
Government (General)	0.029	0.611	0.069	0.07	0.383	2.342	0.228	1.568	3.007	1.083	1.136	1.187	11.713
Journalists & Media	0.002	0.15	0.012	0.01	0.069	0.375	0.067	0.321	0.241	0.091	0.124	0.161	1.623
Maritime	0.001	0.021	0.001	0.002	0	0.031	0.004	0.017	0.027	0.041	0.035	0.013	0.193
Military	0.008	1.687	0.05	0.022	0.625	5.102	0.116	0.974	3.135	1.259	1.663	0.761	15.402
NGO	0.001	0.014	0.006	0.001	0.021	0.074	0.019	0.032	0.163	0.024	0.154	0.026	0.535
Other	0	0.002	0	0	0.003	0.046	0.001	0.001	0.009	0.005	0.003	0.006	0.076
Police	0.018	0.321	0.042	0.054	0.482	3.794	0.129	1.242	4.662	0.742	0.824	1.177	13.487
Private Citizens & Property	0.02	0.826	0.043	0.053	0.474	8.397	0.258	1.735	5.774	1.377	3.147	1.844	23.948
Religious Figures/Institutions	0.015	0.047	0.004	0.018	0.097	0.695	0.094	0.192	0.628	0.188	0.296	0.17	2.444
Telecommunication	0.001	0.083	0.003	0.003	0.009	0.039	0.006	0.079	0.167	0.086	0.036	0.046	0.558
Terrorists/Non-State Militia	0	0.019	0.005	0.003	0.01	0.936	0.004	0.043	0.384	0.043	0.079	0.147	1.673
Tourists	0.001	0.009	0.001	0.002	0.005	0.08	0.007	0.021	0.028	0.019	0.019	0.051	0.243
Transportation	0.006	0.223	0.019	0.081	0.162	0.653	0.021	0.594	1.162	0.254	0.319	0.247	3.741
Unknown	0.001	0.046	0.006	0.003	0.07	1.189	0.008	0.091	1.322	0.161	0.143	0.207	3.247
Utilities	0.002	0.675	0.005	0.003	0.047	0.447	0.057	1.062	0.44	0.21	0.261	0.105	3.314
Violent Political Party	0.002	0.017	0	0.001	0.004	0.219	0.009	0.015	0.609	0.007	0.113	0.032	1.028
Total	0.158	5.691	0.311	0.442	2.829	27.781	1.902	10.445	24.754	6.875	9.662	9.159	100.009

Based on the summary, we can state the following facts.

• 11.37% of the terrorist attacks in all of its varieties are targeted to business. which includes attacks carried out against corporate offices or employees of firms like mining companies, or oil corporations. Furthermore, includes attacks conducted on business people or corporate officers. Included in this value as well are hospitals and chambers of commerce and cooperatives.

• 11.71% of the terrorist attacks are targeted to any government member,member, former members, including members of political parties in official capacities, their convoys, or events sponsored by political parties; political movements or a government sponsored institution where the attack is expressly carried out to harm the government.

• 23.94% of the attacks in the world includes attacks on individuals, the public in general or attacks in public areas including markets, commercial streets, busy intersections and pedestrian malls.

• 15% of the attacks in the world includes attacks against military units, patrols, barracks, convoys, jeeps, and aircraft.

• 13% of the attacks in the world are targeted to members of the police force or police installations; including police boxes, patrols headquarters, academies, cars, checkpoints, etc. Includes attacks against jails or prison facilities, or jail or prison staff or guards.

• Other type of targets such as Abortion Related, Airports & Aircraft, Educational Institution, Food or Water Supply, Government (Diplomatic), Journalists & Media, Maritime, etc…, are less likely to be targeted with probabilities less than 4%.

With the regards to the three most active regions of the world, Middle East & North Africa have the higer number of incidents with 27.78% of the attacks that happen in the world, South Asia follows with a 27.7% and finally South America with a 10.44%.

Type of Attacks

This Section shows the general method of attack class of tactics used. It consists of nine categories, which are defined below. Only one attack type is recorded for each incident unless the attack is comprised of a sequence of events. Attacks are also recorded based on the goal, not the method used. For example, if an assassination is carried out through the use of an explosive, the Attack Type is coded as Assassination, not Bombing/Explosion.

atk_type <- terror_dt %>%
  select(attacktype1_txt,region_txt) %>%
  group_by(attacktype1_txt,region_txt) %>% tally(name = "attack_no")

atk_type %>% ggplot(aes(y = attack_no,x = fct_reorder(region_txt,attack_no), fill=attacktype1_txt))+geom_bar(stat = "identity"  )+theme(axis.text.x = element_text(angle = 40,vjust = 0.8,hjust = 1))+facet_wrap(.~attacktype1_txt,nrow = 3,scales = "free_x")+coord_flip()+ggtitle("Type of Attacks by Region")+xlab("Number of Attacks")+ylab("Number of Deaths")+theme(legend.position = "none")

Table summary by region

atk_type_summary <- atk_type %>% spread(key = attacktype1_txt,value = attack_no)
atk_type_summary %>%
  kable(col.names = c( "Region","Armed Assault", "Assassination","Bombing/Explosion","Facility/Infrastructure Attack" ,  "Hijacking","Hostage Taking (Barricade Incident)", "Hostage Taking (Kidnapping)", "Unarmed Assault","Unknown"    ),caption = "Table 6: Summary of incidents by Region") %>% 
 kable_styling(position = "left",full_width = F,bootstrap_options = "striped",font_size = 13) %>% scroll_box(width = "100%")

Table 6: Summary of incidents by Region

Region	Armed Assault	Assassination	Bombing/Explosion	Facility/Infrastructure Attack	Hijacking	Hostage Taking (Barricade Incident)	Hostage Taking (Kidnapping)	Unarmed Assault	Unknown
Australasia & Oceania	51	36	75	71	3	6	13	11	16
Central America & Caribbean	4361	1254	3239	403	26	187	501	19	354
Central Asia	116	115	235	20	8	2	45	5	17
East Asia	117	55	330	200	18	3	14	42	23
Eastern Europe	1274	400	2766	260	26	21	220	62	115
Middle East & North Africa	9273	4206	30908	1115	138	100	2666	177	1891
North America	448	255	1534	906	18	67	123	73	32
South America	3875	2745	9039	803	67	234	1414	47	754
South Asia	11404	4301	21246	2189	93	120	3277	323	2021
Southeast Asia	4022	1369	4818	948	59	67	744	25	433
Sub-Saharan Africa	6004	1638	5557	810	136	95	1872	83	1355
Western Europe	1724	2938	8508	2631	67	89	269	148	265

Conclusion

Based on the data we can conclude that incidence of terrorism around the world have, in overall, increased with the pass of the years, however, such changes are relative to the regions where the incidences took places since increases in the terrorism incidents in certaing regions of the world despite high, does not compare to other regions of the world.

As a matter of fact, the higer total number of victims around the world are located in the Middle East & North Africa which represents a 33.41% of the world wide victims whereas the Sub-Saharan Africa region and South America represents the 19.04% and the 7% of the world wide victims respectively. On the other hand, contrary to popular believes, regions like North America
and Europe (West and East) represent only a 1.2% and a 3.4% of all the incidents around the world.

Since the data is continously modified in order to increase the accuracy of the information about certain incidents, in the future I expect the get more data in order to calculate better proportion estimates for the incidents that happened between 1970 and 2017 and possibly 2018.