In Chapter 5, you saw that sometimes, instead of manipulating what our study participants see and experience, we use non-experimental methods; we just observe our participants in their natural environments without influencing them at all. In these cases, we are using descriptive research methods. The lack of control in descriptive research methods means that we do not have the ability to establish cause and effect.
The correlational research approach, a popular research method, is another non-experimental research approach. In correlational research, we measure different variables to see whether there is a relationship between or among them. In other words, you would have a relationship between measured variables if they vary systematically together (as intelligence rises, does academic performance rise as well?). I'll explain more about this below. Before we go on, remember that since correlational research is another descriptive research technique, we cannot use it to investigate whether one variable causes another.
The Basic Differences between Experimentation and Correlational Research
Before we talk about the details of correlational research, I'd like to explain the basic methodological differences between experimentation and correlational research.
Experimentation
Consider the simplest kind of experiment. It has one independent variable with two levels represented by two groups with different people in each group. Here are the steps we take in a simple experiment:
We randomly assign participants to the two groups.
We manipulate (vary) an independent variable. Say we give one group 2 oz. of caffeine and the other group 10 oz.
We measure a dependent variable, say performance on a math test.
There are indeed advantages to doing experimental research; number one is the ability to establish cause and effect (as explained in Chapter 1). Unfortunately, there are times when we cannot do highly controlled experiments due to practical or ethical concerns. Take the following example:
• Does illegal drug use during their college years increase students’ risk of having difficulties in college?
We cannot randomly put study participants into two groups, and give one group illegal drugs and not give illegal drugs to the other group; this would be unethical. Whether a student takes illegal drugs is not a decision that is up to experimenters. To investigate these types of questions, researchers instead often turn to correlational research.