How to Experiment Well (Lesson 4.6)
Chapter 4 - Day 8
Describe how to randomly assign treatments in an experiment using slips of paper or technology.
Explain the purpose of random assignment in an experiment.
Identify other sources of variability in an experiment and explain the benefits of keeping these variables the same for all experimental units.
We started Chapter 4 with an intro activity that posed the question “Is anchored putting better?”. In this lesson, we start to formalize some of the thinking that students did in the intro activity, with more formalization coming in Lesson 4.7. Remind students of this context, and show them the putters again if you have them (if you don’t have them yet, talk to your golf coach or head to Play-it-Again Sports as you will need putters for the finale activity at the end of the Chapter).
For questions #2 and #3 students are asked to explain how to do random assignment for an experiment. Many of them will make the connection that the process of random assignment is very similar to selecting a simple random sample. This is a valid connection and try to help them to remember what ideas were important when taking a simple random sample (equal sized slips of paper, shuffle slips of paper, ignore repeats when using a random number generator).
When debriefing #4, we don’t expect students to say “roughly equivalent”. This is language that we introduce after allowing them to explain their answer. While it is likely that students will use golf skill as the confounding variable that we are trying to prevent, consider presenting some other possible confounding variable (like whether a student is right-handed or left-handed). Help students realize that random assignment should roughly balance the two groups with right-handed and left-handed students. In fact, random assignment should create two groups that are roughly equivalent with respect to every variable that might affect the results.
For debriefing #5, try to use a specific variable to present each of the two benefits of keeping other variables the same. If we didn’t use the same ball for every student in the experiment, we wouldn’t know whether it was the ball or the putter that affects the results (random assignment helps to prevent confounding). If we let each student putt from a different distance, we would end up with some putts far from the hole and some close to the hole just because of the distance, not because of the putter (random assignment reduces variability).
The next Lesson 4.7 will continue within this same putting context. In this lesson, students will draw an outline for this experiment, discuss statistical significance, and do a simulation.