Download presentation
Presentation is loading. Please wait.
Published byAlexandra Davidson Modified over 5 years ago
1
Warm Up One day Dr. Lutze decides he wants to randomly assign his first period students to 2 groups for an experiment on time management. In order to save time he decides he will flip a coin as each student enters the room; if the coin is heads the student will be assigned to group 1, if the coin is tails the students will be assigned to group 2. There are 34 students in the class so Dr. Lutze will continue the coin flipping until one group has 17 students. At that point he will assign all the remaining students to the other group. Is this an effective method for random assignment? Explain why or why not.
2
Experimental Design – Matched Pairs
Match the experimental units in pairs that are similar or identical to account for lurking variables. Matched pair designs compare 2 treatments only. Example: Compare the yield of two seed types on a farm. Create matched pairs so that the two seed types are always planted near each other. Controls for variation in sunlight, water, pests, etc. Randomize which of the pair receives which treatment (left/right, first/second). Individuals can be paired with themselves (before/after experiments). Data is the difference between the 2 measurements.
3
Experimental Design – Matched Pairs
Imagine an experiment to test the effect of using a calculator on an Algebra 1 test. Each subject is paired with another student who has a very similar grade in the class before taking the test Calculator Random Assignment 2 students 99%, 98% Calculate Difference No Calculator Calculator 2 students 97%, 96% Random Assignment Subjects Calculate Difference No Calculator Calculator 2 students 61%, 60% Random Assignment Calculate Difference No Calculator Matched pair design eliminates the factor of the students’ grades from influencing the response variable.
4
Experiment – Effect on Your Pulse Rate of a Fast Walk
Each subject (you) will measure his/her pulse rate before walking and after walking. - Measure your resting pulse for 30 seconds and double the value to get beats per minute. - You will walk quickly for 1 lap around the portables and the basketball hoops. - After returning to class measure your pulse for 30 seconds and double the value to get beats per minute. - Write your data in your notebook as a matched pair: (Before walking, After walking)
5
Randomized Design – Effect on Your Pulse Rate of a Fast Walk
1) Every student will be randomly assigned to the resting or walking group. 2) If you get a red card you are in the resting group – write your resting pulse on the board. 3) If you get a black card you are in the walking group – write your walking pulse on the board. 4) Determine the mean and standard deviation for resting and walking pulse separately. Is this significant data that a fast walk increases a person’s pulse rate?
6
Matched Pairs Design – Effect on Your Pulse Rate of a Fast Walk
1) Write your data on the board as a matched pair: (Before walking, After walking) 2) Enter the before walking pulses into L1. 3) Enter the after walking pulses into L2. 4) Enter L2 – L1 into L3. 5) Find the mean and standard deviation for the data set in L3. Is this significant data that a fast walk increases a person’s pulse rate?
7
Experiment – Effect on Your Pulse Rate of a Fast Walk
Which experimental design (matched pairs vs. randomized) gave a more significant result? Why?
Similar presentations
© 2025 SlidePlayer.com. Inc.
All rights reserved.