This tutorial will explain matched-pair design experiments by examining the characteristics and examples of:
Finding subjects/participants in pairs as closely matched as possible is the basis of the matched-pair design experiments.
IN CONTEXT
Suppose that you have a tire company that's considering rolling out a new type of rubber for its bicycle tires. There are 300 bicycles available. In a completely randomized design, you would place the numbers 1 - 300 in a hat. Bikers that pull numbers 1 -150 would receive old rubber tires, and the 151- 300 would receive the new rubber tires. They won’t necessarily know who's getting which tires.
But what if the 300 riders don't all ride the same way or equally as often? What do you do then? How do you create two groups that are roughly the same, with the exception of the bicycle tires?
One way to do it is with a matched-pair design. You could still put the numbers 1 - 300 in a hat. The only difference is that the people who pull out 1- 150 would get both the old and the new. They would put the old in the front and the new rubber tire in the back. And the people who pulled out 151 - 300 would get the new rubber tire in the front and the old one in the back.
So there's still some randomization going on. The only difference is that every bikers will get one old tire and one new tire. This will allow you to compare the tread wear for each bike, because the front and rear tire get worn somewhat equally. It won't matter how much the biker rides or where.
Matched-pairs design involves matching subjects into groups of two that are as similar as possible with respect to any variable that may affect the outcome.
Typically, in a matched-pair design, each subject is assigned to both groups instead of one, then randomly assigned the order in which treatments are applied. And each participant then counts as his or her own matched pair. This design essentially compares someone to themselves.
However, a matched pairs design can work like this:
IN CONTEXT
There are 20 participants for an experiment for a flu vaccine. Gender and age may play a role in how well this treatment works. Groups of two are created; each group is as similar as possible with respect to any variable that may affect the outcome.
There are 10 men and 10 women-- all different ages. Participants will be listed by gender. So participant 1, 3, 4, 8, 10, 11, 12, 16, 18, and 20 are the males. The rest are females.
Age is suspected to also play a role in effectiveness, so within the male category, two ages that that are closest together-- 24 and 25 - - are chosen. So participants 1 and 8 will form a matched pair. Participants 10 & 12, 4 & 3, 20 & 11, and 16 & 18 are also matched pairs due to similarly aged males. The same criteria is applied for similarly aged females.
Now, to continue the experiment, one of the two in the pair is randomly assigned to receive the flu vaccine and the other one will be assigned to the control group.
In a matched-pair design, two numbers whose characteristics are very similar are paired, then each one is sent to a different group. When applying matched-pair design, typically, each subject is assigned to both groups instead of one, as was the case with the bicycle tires situation. Matched-pairs designs are often done by assigning both treatments to every participant, which is commonly used in the matched-pairs design.
Good luck!
Source: Adapted from Sophia tutorial by Jonathan Osters.
An experimental design where two subjects who are similar with respect to variables that could affect the outcome of the experiment are paired together, then one of them is assigned to one treatment and one is assigned to the other. This can also be done by assigning each subject to both treatments, where each subjects acts as their own matched-pair.