AP Statistics Handout: Lesson 6.
1
Topics: introduction to sampling distributions (through simulation)
What is Gerrymandering?
1. As a class, watch the following video (direct link - [Link]
2. To get a real sense of how gerrymandering works, try it out yourself! Use this simulation to
practice gerrymandering: [Link] (Stop after the 6th challenge)
At this point, you might be wondering: “Why can’t we just make everything proportional? For
example, if 60% of people in a state support Republicans, let’s just make 6 out of their 10
representatives Republican!” Unfortunately, it’s not so simple…
3. Here is Idaho, with vote patterns broken down by census tracts of
roughly equal population size. Idaho gets 2 House representatives. In
a typical election, Democrats earn 34% and Republicans earn 66% of
major party votes. Both Idaho representatives are almost always
Republicans. (source)
a) To give the most proportional representation, how many of the 2
representatives should be Republican? How many Democrat?
b) In order to give one seat to a Democrat, describe the district you’d
have to draw in Idaho. Does it make sense to have a district that looks
like this? Why or why not?
Material adapted from the Skew The Script curriculum ([Link])
2
The Statistician’s Answer: Simulation
Between balancing the need for party representation (proportionality) and the need for
geographic representation (no weird shapes), the Supreme Court has struggled to set a
consistent standard for evaluating maps. So, in 2019, a group of mathematicians got together
and wrote a Supreme Court brief that proposed a new simulation method for map evaluation:
● Have a computer draw a bunch of randomly simulated maps under two rules: a) districts
have equal population sizes, b) no weird shapes.
● Using prior election data, estimate how well Democrats and Republicans would do in
each of the randomly drawn maps.
● See if the map drawn by the actual state legislature is unusual (in terms of projected
election outcomes) compared to the randomly simulated maps.
By having a computer (with no internal party preference and with rules for shapes) draw
random maps, we balance the needs for both proportionality and geography!
4. Open up the redistricting simulation applet ([Link]/Gerrymandering).
Perform an analysis for Virginia. During redistricting, Virginia's state legislature was controlled
by Republicans. Did the legislature draw an unfair map? Support your answer with evidence.
5. Using the applet, perform an analysis for Oregon. During redistricting, the state legislature
was controlled by Democrats. Did they draw an unfair map? Support your answer.
6. Using the applet, perform an analysis and comment on the fairness of…
a) Illinois (Democrat-controlled legislature) and Texas (Republican-controlled legislature)
b) Your own state (note: states with 1 house member don’t go through redistricting).
Material adapted from the Skew The Script curriculum ([Link])
3
Lesson 6.1 Discussion
In its most recent landmark decision on gerrymandering (Rucho v. Common Cause), the
Supreme Court abdicated to the legislature, saying that the courts could not make decisions on
"political questions outside the remit of these courts." In other words, if change is to happen,
state legislatures or Congress have to pass laws to limit gerrymandering.
Discussion Question: Write such a law, using the simulation method described in this lesson.
Specifically, fill in the blank: “If only ___% of simulations are more extreme than your map, it
should be rejected.” What number would you choose for the blank? Why?
Lesson 6.1 Practice
1) Your teacher gives students an opportunity to draw a marble out of a well-mixed bag, in
hopes of picking a green marble. A green marble means you get a homework pass. The teacher
claims that 20% of the marbles in the bag are green. Your class has 20 students in it.
One-by-one, without looking, each student selects 1 marble from the bag. After drawing their
marble, each student puts it back in the bag.
a) If your teacher is telling the truth about your chances of winning a homework pass by
picking a green marble, how many students would you expect to win in your class of 20?
b) Imagine that in your class of 20 students, no one picks a green marble. Students start to
wonder if fewer than 20% of the marbles are actually green. Without drawing more
marbles, what could the students do to evaluate how unlikely it was to get 0 green
marbles in 20 draws?
Material adapted from the Skew The Script curriculum ([Link])
4
The dotplot below shows results from 100 random computer simulations of n = 20 students
drawing marbles from a bag, in which 20% of the marbles in the bag are green. The x-axis
displays the percent of students who got a green marble (out of 20 students). Each dot
represents a simulation result.
Created using [Link]
c) In your class, you observed nobody getting a green marble. Using the results of the
simulation above, was this outcome surprising? Would you start to doubt the teacher’s
claim that 20% of the marbles in the bag are green? Explain your thinking.
Further Practice
Teachers: We recommend providing additional practice exercises from your textbook or prior AP
Stats exams. The following textbook sections and AP exam questions are aligned to this lesson.
● The Practice of Statistics (AP Edition), 4th-6th editions: section 7.1
● Stats: Modeling the World (AP Edition), 4th/5th editions: ch 17, 3rd edition: ch 18
● Statistics: Learning from Data (AP Edition), 2nd edition: section 8.1
● Advanced High School Statistics, section 4.1
Thank you to…
● Max Cembalest & Chris Olsen, for their help in writing the applet and lesson.
● The creators of the redist R package, whose work powers the applet.
● Jordan Ellenberg, whose book chapter inspired this lesson.
Material adapted from the Skew The Script curriculum ([Link])