8.3: Between Groups Designs and Matched Group Designs

Learning Objectives

1. Explain the difference between group designs and matched group designs.
2. Develop ideas about the benefits and disadvantages of between groups and matched group designs.

In this section, we look at two different ways to design an experiment by randomly assigning individual participants into IV conditions so that the IV conditions are expected to have initial equivalence: between groups designs and matched group designs. The following section will focus on within groups designs, with repeated measures designs being the easiest to understand.

Between Group Design

In a between groups design, each participant in the experiment is randomly assigned into only one condition. For example, a researcher with a sample of 100 participants might assign half of them to limit their social media use and the other half of the participants are not asked to change their social media usage by having each participant flip a coin; heads means that they should limit their social media use while tails means that they do not. Or a researcher with a sample of 90 people with severe agoraphobia (fear of open spaces) might randomly assign 30 of them to receive each of three different treatments for that disorder. The randomness of the assignment procedure into the IV group to ensure initial equivalence between the groups is an important characteristic of this process; it is essential in a between groups design that the researcher assigns participants to conditions so that the different groups are, on average, highly similar to each other. Those in the limited social media use condition and those in the unlimited social media condition, for example, should include a similar proportion of each gender, and they should have similar average levels of anxiety, similar average age, and so on. The random assignment used in hope of controlling these extraneous participant variables across conditions so that they do not become confounding variables.

Matched Groups

An alternative to simple random assignment of participants to conditions is the use of a matched groups design. Using this design, participants in the various conditions are matched on a variable that is not the IV or DV, but might be expected to affect the DV, prior the assignment into an IV condition and manipulation of the independent variable. Then, the matched groups are randomly assigned into the various IV conditions, ensuring that there is initial equivalence between the IV conditions on at least that one variable.

Figure \(\PageIndex{1}\): Basketball Through Hoop by Airman 1st Class Kerelin Molina via https://commons.wikimedia.org/wiki/F...rough_hoop.jpg . This image or file is a work of a U.S. Air Force Airman or employee, taken or made as part of that person's official duties. As a work of the U.S. federal government, the image or file is in the public domain in the United States

This is a confusing concept for students to grasp, here's an example: Imagine that your Physical Education (PE) instructor wanted to know what kind of practice most helps to improve making lay-ups in basketball. The PE instructor thinks that practicing shooting from the key would help, but also notices that many players struggle with lay-ups because they dribble the basketball poorly. So, the instructor's two IV conditions will focus on shooting from the key or practicing dribbling. But one other factor might affect lay-up skill, which is height. Imagine your most recent PE class lined up from tallest to shortest. Then, pairs of students were made based on their similar heights; the two tallest students would be paired, then the next two tallest students, and so on until the two shortest students are paired. Then, each pair (who have similar heights) would use a random procedure, such as playing rock/paper/scissors (sometimes called Rochambeau, Roshambo, Ro-sham-bo, Bato Bato Pik, or Jak-en-poy, Wikipedia, 2025) to determine IV group assignment. The winner could be assigned to the group to practice shooting from the key, and the rock/paper/scissors loser could be assigned to the group that practiced dribbling. After all pairs are assigned, there would be two groups: a group practicing shooting from the key and a group practicing dribbling. The key is that the average height of each group would be similar; there would be tall students in the group practicing shooting from the key and there would be tall students in the group who are practicing dribbling. There would also be shorter students in each group. Because the students were matched on height and then randomly assigned into the IV condition from that matched pair, we are very confident that the groups will be initially equivalent on height; height will not be an extraneous variable that could affect scores on the DV of making lay-ups. And because we still used random assignment, we should have initial equivalence on all other variables. At the end of the study, if the PE instructor found that number of lay-ups made was statistically higher for students from one type of practice, we would know that the results were not influenced by height; the matching guarantees that the matched variables will not be confounded across the experimental conditions.. However, we can't be sure of this unless we measure each variable that we are concerned about.

---

References

Wikipedia (2025, July 2). Rock paper scissors. https://en.wikipedia.org/wiki/Rock_paper_scissors