Homework 4


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.


Skills used:

  • grouped summary counts
  • grouped summary statistics
  • plotting
  • creating new variables
  • sorting

My awkward Google Sheets video tutorials are here

US Army Enlistment

Let’s look at this data which gives us county-level information on military service rates across the USA (variable dictionary below).

Variable Description
fip ID for county
state Name of state
county Name of county
base Is there a military base in this county? (1 = yes, 0 = no)
ms_pre_2001 Average num. of people in county who served in military (pre-2001)
ms_post_2001 Average num. of people in county who served in military (post-2001)
pop2010 County population
black_2010 Percent of county who is Black
hispanic_2010 Percent of county who is Hispanic
hs_grad_2010 Percent of county with a high school degree
median_household_income_2010 Median household income in county
unemployment_rate_2010 Unemployment rate in county
poverty_2010 Poverty rate in county
  1. Which county had the highest post-2001 military service?

  2. A big problem with (1): that county might have the highest military service because many people enlisted there, or it might just be that there’s a lot of people who live in that county and that’s inflating the military service variable. Create a per-capita enlistment variable called rate that divides post-2001 enlistment by county population. Which county has the highest enlistment rate?

  3. What is the relationship between a county’s unemployment rate and its a military service rate (new variable you constructed in question 2)? Plot them together (x-axis = unemployment rate, y-axis = service rate) and add a trend line so you can see the general trend. Roughly, what is that trend? Save the plot to upload. (note: once you have the scatterplot up, you can add a trend line by going to Customize -> Series -> scroll down to “trendline” checkbox)

  4. Look at the military bases variable. How many military bases does the state with the most bases have?

iCasualties data

Let’s look at this data on virtually all American casualties in the Iraq and Afghanistan wars (variable dictionary below). Each row here is a soldier who died in one of these two wars.

  1. Pick a random row/observation — look up the soldier online. What happened to this person?

  2. About what percent of casualties are the result of non-hostile actions (things like accidents, friendly fire, etc.)? Look at the source variable.

  3. What was the most common cause of death for US soldiers?

  4. do anything else with either dataset. Tell me what you did and what you found.

Variable Description
date Date of death
name Name of soldier
rank Rank of soldier
nationality Nationality of soldier
branch Branch of military
age Age at death
country Country (Iraq or Afghanistan)
province Province where soldier died
where City/locale where soldier died
source Source of death (hostile or non-hostile)
cause Cause of death
state Home state of soldier
city Home city/town of soldier
state_pop_2010 Population in state (2010)