Homework 2


Skills used:

  • grouped summary counts
  • grouped summary statistics
  • plotting

My awkward Google Sheets video tutorials are here


Answer the questions (below), and keep track of your answers somewhere (a notepad?). You will input your answers into Canvas. Figure out the answers BEFORE opening Canvas, otherwise you’ll run out of time.


You’re going to work with a random sample from the Uppsala Conflict Data Program at The Department of Peace and Conflict Research, Uppsala University in Oslo, Norway. The dataset is called the Geo-referenced Event Dataset (GED) which provides detailed geo-referenced information on violent events around the world.

Variable Description
id Unique event identifer
year Year event began
month Month event began
day Day event began
country Country of event
type_of_violence UCDP violence type
conflict_name Name of conflict
side_a Side A
side_b Side B
source_article Article event came from
latitude Latitude
longitude Longitude
low_death_estimate Low estimate of number of deaths
high_death_estimate High estimate of number of deaths
best_death_estimate Best estimate


Using the dataset, answer the following questions:

All of these events are geo-located, meaning they have latitude and longitude coordinates that identify their location in the world. Find a violent event that took place on your birthday (if there are none, try adding/subtracting a day, or month, until you find one).

  1. Use the longitude and latitude coordinates of that event to find it on Google Maps. Using either street view or, if street view is not available, whatever images you can find near the location of the event, take a screencap and save the image (you will upload to Canvas).

  2. Using the information in the other columns, write 3-4 sentences in plain English about: a) the conflict in which that event took place; b) the violent event itself, just describe what exactly happened (as best as you can tell: who did what to whom, and why).

  3. Based on the images you captured, very briefly describe what you’re seeing in the image. What do you see? What does it look like?

Afghanistan is famously known for having a “fighting season”, in which combat is typically heaviest in a particular time of the year.

  1. Count the number of violent events that took place in Afghanistan per month (grouped summary count). Note: You can use the id column to count the events. Make a barplot/chartplot/bar chart/column chart of the result and save it as an image to upload (plot).

  2. Look up the Afghanistan fighting season on Wikipedia, read about it, and compare your results to what you see on the barplot/bar chart/column chart in 2-3 sentences.

Finally, looking at the whole dataset (not just Afghanistan):

  1. The UCDP categorizes violent events into three violent event types. Look them up: what are those three types? What does each mean?

  2. Calculate the average number of deaths in a violent event (using the “best” estimate variable), per type of violent event (grouped summary statistics). Which type of violent event produces the most deaths, on average?

  3. Calculate the maximum number of deaths in a violent event (using “best” estimate) per country (grouped summary statistics). In which country were the most deaths recorded in a single violent event?