ETC1010-5510: Introduction to Data Analysis
Melisa
Load all the libraries that you need here
library(tidyverse)
library(dplyr)
library(ggplot2)Read in the data
Question 1: Display the first 10 rows of the data set (1pt). Hint: Check ?head in your R console
head(terror,10) #Run in R console## # A tibble: 10 × 135
## eventid iyear imonth iday approxdate extended resolution country
## <dbl> <dbl> <dbl> <dbl> <chr> <dbl> <dttm> <dbl>
## 1 1.97e11 1970 7 2 <NA> 0 NA 58
## 2 1.97e11 1970 0 0 <NA> 0 NA 130
## 3 1.97e11 1970 1 0 <NA> 0 NA 160
## 4 1.97e11 1970 1 0 <NA> 0 NA 78
## 5 1.97e11 1970 1 0 <NA> 0 NA 101
## 6 1.97e11 1970 1 1 <NA> 0 NA 217
## 7 1.97e11 1970 1 2 <NA> 0 NA 218
## 8 1.97e11 1970 1 2 <NA> 0 NA 217
## 9 1.97e11 1970 1 2 <NA> 0 NA 217
## 10 1.97e11 1970 1 3 <NA> 0 NA 217
## # ℹ 127 more variables: country_txt <chr>, region <dbl>, region_txt <chr>,
## # provstate <chr>, city <chr>, latitude <dbl>, longitude <dbl>,
## # specificity <dbl>, vicinity <dbl>, location <chr>, summary <chr>,
## # crit1 <dbl>, crit2 <dbl>, crit3 <dbl>, doubtterr <dbl>, alternative <dbl>,
## # alternative_txt <chr>, multiple <dbl>, success <dbl>, suicide <dbl>,
## # attacktype1 <dbl>, attacktype1_txt <chr>, attacktype2 <dbl>,
## # attacktype2_txt <chr>, attacktype3 <lgl>, attacktype3_txt <lgl>, …Question 2: How many observations and variables does the data set terror have (1pt)? Use inline code to complete the sentence below (2pts)
nrow(terror)## [1] 209706ncol(terror)## [1] 135Data terror contains 209706 record rows and 135 columns
Question 3: What is the name of the 3rd variable in this data set (2pts)? Use R commands to answer this question.
variables<- names(terror)
variables[3]## [1] "imonth"Question 4: Using the terror data set, rename the first variable to event_id and save this new data frame as terror (2pts). Display the first 4 rows and 9 columns corresponding to the country “France” (1pt).
terror<-rename(terror,"event_id"="eventid")terror1<-terror %>%
dplyr::filter(country_txt == "France") %>% head(4)
terror1 %>% select("event_id","iyear","imonth","iday","approxdate","extended","resolution","country","country_txt") ## # A tibble: 4 × 9
## event_id iyear imonth iday approxdate extended resolution country
## <dbl> <dbl> <dbl> <dbl> <chr> <dbl> <dttm> <dbl>
## 1 1.97e11 1972 5 25 <NA> 0 NA 69
## 2 1.97e11 1972 5 25 <NA> 0 NA 69
## 3 1.97e11 1972 5 25 <NA> 0 NA 69
## 4 1.97e11 1972 5 25 <NA> 0 NA 69
## # ℹ 1 more variable: country_txt <chr>Question 5: Using the terror data set, and the function
filter extract all data between the years of 1995 and 2011.
Call this new dataset terror_short.(4pt)
terror_short<-terror %>%
dplyr::filter(iyear >= 1995 & iyear <= 2011 )
terror_short## # A tibble: 46,626 × 135
## event_id iyear imonth iday approxdate extended resolution country
## <dbl> <dbl> <dbl> <dbl> <chr> <dbl> <dttm> <dbl>
## 1 2.00e11 1995 1 0 <NA> 0 NA 217
## 2 2.00e11 1995 1 0 <NA> 0 NA 217
## 3 2.00e11 1995 1 0 <NA> 0 NA 217
## 4 2.00e11 1995 1 1 <NA> 0 NA 202
## 5 2.00e11 1995 1 1 <NA> 0 NA 209
## 6 2.00e11 1995 1 1 <NA> 0 NA 55
## 7 2.00e11 1995 1 1 <NA> 0 NA 55
## 8 2.00e11 1995 1 1 <NA> 0 NA 55
## 9 2.00e11 1995 1 1 <NA> 0 NA 55
## 10 2.00e11 1995 1 1 <NA> 0 NA 83
## # ℹ 46,616 more rows
## # ℹ 127 more variables: country_txt <chr>, region <dbl>, region_txt <chr>,
## # provstate <chr>, city <chr>, latitude <dbl>, longitude <dbl>,
## # specificity <dbl>, vicinity <dbl>, location <chr>, summary <chr>,
## # crit1 <dbl>, crit2 <dbl>, crit3 <dbl>, doubtterr <dbl>, alternative <dbl>,
## # alternative_txt <chr>, multiple <dbl>, success <dbl>, suicide <dbl>,
## # attacktype1 <dbl>, attacktype1_txt <chr>, attacktype2 <dbl>, …Question 6: How many variables and observations are in the
terror_short data frame during these years (2pts)? Hint: you
can use count or nrow to complete this.
nrow(terror_short)## [1] 46626ncol(terror_short)## [1] 135Data terror contains 46626 record rows and 135 columns
Question 7: Use terror_short dataframe to answer the following questions.
Question 7.1: Since 1995, which countries have experienced the most terrorist attacks. Display the five countries with the highest frequency of attacks. (5pt)
terror_short %>% group_by(country_txt) %>% summarise(ncount=n()) %>% arrange(-ncount) %>% head(5) ## # A tibble: 5 × 2
## country_txt ncount
## <chr> <int>
## 1 Iraq 7742
## 2 Pakistan 4833
## 3 India 4638
## 4 Afghanistan 2935
## 5 Colombia 2585Question 7.2: In Western Europe, which county has experienced the most attacks.(2pt)
terror_short %>% filter(region_txt=="Western Europe")%>% group_by(country_txt) %>% summarise(ncount=n()) %>% arrange(-ncount) %>% head(1) ## # A tibble: 1 × 2
## country_txt ncount
## <chr> <int>
## 1 France 771Question 7.3: What are the three most common type of attack in Eastern Europe? Use the variable ‘attacktype1_txt’ to answer this question. (2pt)
terror_short %>% filter(region_txt=="Eastern Europe")%>% group_by(attacktype1_txt) %>% summarise(ncount=n()) %>% arrange(-ncount) %>% head(3) ## # A tibble: 3 × 2
## attacktype1_txt ncount
## <chr> <int>
## 1 Bombing/Explosion 1370
## 2 Armed Assault 624
## 3 Assassination 259Question 8: Use terror_short dataframe to answer the following questions.
Question 8.1: Calculate the number of attacks per month across all years. Display the month with the highest number of attacks. (4pt)
terror_each_month <- terror_short %>% group_by(iyear,imonth)%>% summarise(number_of_attack=n()) %>% arrange(-number_of_attack)## `summarise()` has grouped output by 'iyear'. You can override using the
## `.groups` argument.terror_each_month## # A tibble: 204 × 3
## # Groups: iyear [17]
## iyear imonth number_of_attack
## <dbl> <dbl> <int>
## 1 2011 11 661
## 2 2007 6 588
## 3 2001 8 568
## 4 2008 7 554
## 5 2007 11 534
## 6 2010 9 528
## 7 2008 6 510
## 8 2008 4 507
## 9 2008 5 500
## 10 2009 7 467
## # ℹ 194 more rowsQuestion 8.2: Calculate the average number of attacks per month across all year, and display the month with the highest average. Which month has the highest variability as measured by standard deviation? (5pt)
terror_each_month %>% group_by(imonth) %>% summarise(mean=mean(number_of_attack),sd=sd(number_of_attack))%>% arrange(-sd)## # A tibble: 12 × 3
## imonth mean sd
## <dbl> <dbl> <dbl>
## 1 11 242. 178.
## 2 6 244. 166.
## 3 8 257. 145.
## 4 4 227. 145.
## 5 5 239. 144.
## 6 7 255. 143.
## 7 12 200. 133.
## 8 9 208. 127.
## 9 2 204. 126.
## 10 10 229. 120.
## 11 3 219. 117.
## 12 1 219. 103.Question 8.3: Use a boxplot to display the frequency of attacks by Month across all years. By visual analysis, state which month has the highest and lowest variability? (2pts)
ggplot(terror_each_month, aes(x =factor(imonth),y=number_of_attack,color = factor(imonth)))+
geom_boxplot()
Question 8.4: Recalling the relationship between boxplots and empirical quantiles, state which month has the highest interquartile range of attacks. (4pts)
based on gg-plot above, the highest interquartile of number of attack is either month June or November
terror_each_month %>% group_by(imonth) %>% summarise(q1=quantile(number_of_attack,0.25),q3=quantile(number_of_attack,0.75))%>% arrange(-q3)## # A tibble: 12 × 3
## imonth q1 q3
## <dbl> <dbl> <dbl>
## 1 6 119 376
## 2 4 104 358
## 3 8 125 347
## 4 7 117 337
## 5 11 107 337
## 6 10 119 326
## 7 5 130 323
## 8 9 105 287
## 9 1 164 282
## 10 3 120 270
## 11 12 83 268
## 12 2 112 267After validating it with the gap between Q3 and Q1, we can take conclusion that the highest number of attack occur on June
Question 8.5: Do you observe any pattern in the frequency of attacks throughout the year? (Hint: Make sure your boxplot is aligned from January to December.) Use two to three sentences to answer this question. (5pt)
** Mostly the number of attacks for each month throughout the year is arround 200 but on selected period, specially on July and Aug, frequency of attack tend experience an increasing. Therefore, median number on these period were high**
R Markdown
This is an R Markdown document. Markdown is a simple formatting syntax for authoring HTML, PDF, and MS Word documents. For more details on using R Markdown see http://rmarkdown.rstudio.com.
When you click the Knit button a document will be generated that includes both content as well as the output of any embedded R code chunks within the document. You can embed an R code chunk like this:
summary(cars)## speed dist
## Min. : 4.0 Min. : 2.00
## 1st Qu.:12.0 1st Qu.: 26.00
## Median :15.0 Median : 36.00
## Mean :15.4 Mean : 42.98
## 3rd Qu.:19.0 3rd Qu.: 56.00
## Max. :25.0 Max. :120.00Including Plots
You can also embed plots, for example:
Note that the echo = FALSE parameter was added to the
code chunk to prevent printing of the R code that generated the
plot.