Chapter 4: Measurement and Units of Analysis
While it is very common to hear the terms independent and dependent variable, extraneous variables are less common, which is surprising because an extraneous variable can destroy the integrity of a research study that claims to show a cause and effect relationship. An extraneous variable is a variable that may compete with the independent variable in explaining the outcome. Remember this, if you are ever interested in identifying cause and effect relationships you must always determine whether there are any extraneous variables you need to worry about. If an extraneous variable really is the reason for an outcome (rather than the IV) then we sometimes like to call it a confounding variable because it has confused or confounded the relationship we are interested in. (see example below)
Suppose we want to determine the effectiveness of new course curriculum for an online research methods class. We want to test how effective the new course curriculum is on student learning, compared to the old course curriculum. We are unable to use random assignment to equate our groups. Instead, we ask one of the college´s most experienced online teachers to use the new online curriculum with one class of online students and the old curriculum with the other class of online students. Imagine that the students taking the new curriculum course (the experimental group) got higher grades than the control group (the old curriculum). Do you see any problems with claiming that the reason for the difference between the two groups is because of the new curriculum? The problem is that there are alternative explanations.
First, perhaps the difference is because the group of students in the new curriculum course were more experienced students, both in terms of age and where they were in their studies (more third year students than first year students). Perhaps the old curriculum class had a higher percentage of students for whom English is not their first language and they struggled with some of the material because of language barriers, which had nothing to do with then old curriculum. In other words, we have a problem, in that there could be alternative explanations for our findings. These alternative explanations are called extraneous variables and they can occur when we do not have random assignation. Indeed, it is very possible that the difference we saw between the two groups was due to other variables (i.e. experience level of students, English language proficiency), rather than the IV (new versus old curriculum).
It is important to note that researchers can and should attempt to control for extraneous variables, as much as possible. This can be done in two ways. The first is by employing standardized procedures. This means that the researcher attempts to ensure that all aspects of the experiment are the same, with the exception of the independent variable. For example, the researchers would use the same method for recruiting participants and they would conduct the experiment in the same setting. They would ensure that they give the same explanation to the participants at the beginning of the study and any feedback at the end of the study in exactly the same way. Any rewards for participation would be offered for all participants in the same manner. They could also ensure that the experiment occurs on the same day of the week (or month), or at the same time of day, and that the lab is kept at a constant temperature, a constant level of brightness, and a constant level of noise (Explore Psychology, 2019).
The second way that a researcher in an experiment can control for extraneous variables is to employ random assignation to reduce the likelihood that characteristics specific to some of the participants have influenced the independent variable. Random assignment means that every person chosen for an experiment has an equal chance of being assigned to either the test group of the control group (Explore Psychology, 2019). Chapter 6 provides more detail on random assignment, and explains the difference between a test group and a control group.