Research is the most basic element of a scientific study. It usually begins with an idea or a thought.
Let’s suppose I was wondering if exposure to sunlight can reduce symptoms of depression. That’s the original thought I have.
The first step to conduct research for this thought would be to convert this into a testable Hypothesis or a supposed explanation or a proposition. I would hypothesize that increased exposure to sunlight will lead to decreased symptoms of depression. This is what you would call a Research (or Alternative) Hypothesis. This basically represent my original thought. It assumes that there is a difference in depression between someone who is exposed to sunlight and someone who isn’t.
Now, since science isn’t based on people’s personal opinions, we would like to have an opposing thought as well. A devil’s advocate, so to speak. The opposite of a research hypothesis is a Null Hypothesis. A null hypothesis contradicts the original thought and the purpose of the research is then to identify, whether you would accept or reject the null hypothesis. If you accept the null, you are automatically rejecting the research hypothesis, and vice versa. In our example, the null hypothesis may be increased exposure to sunlight will not lead to decreased symptoms of depression.
After you have formulated your hypotheses, you would need to figure out what type of research design is more appropriate for your study. We will discuss this later.
Within a research study, there are often several Variables. A variable can be an attribute like gender, a behavior like sunlight exposure, or a group of symptoms within a disorder like level of depression. It’s called a variable because it can vary from one person/condition to another. In our example, there are two variables: level of depression and exposure to sunlight.
Typically, variables can be Dependent or Independent. Just as the name suggests, a Dependent Variable “depends” on another variable (usually the independent variable). Since an Independent Variable can be changed or manipulated, it will affect the dependent variable. So, in my proposed hypothesis, the level of depression “depends” on how much exposure someone has to sunlight. Therefore, level of depression is a dependent variable and exposure to sunlight is an independent variable.
Other words used to describe the dependent variable are Criterion, intervention, and treatment. Similarly, the independent variable is sometimes referred to as Predictor or the outcome.
Sometimes a third variable affects the relationship between the dependent and the independent variable. This third variable can be a Mediator or a Moderator variable. A mediator variable is responsible for the relationship between the dependent and the independent variable. The independent variable exerts its effects on the mediator variable, which then exerts its effects on the dependent variable. In our example, if exposure to sunlight increases the levels of vitamin D in your body, which, in turn, improves your mood, then the level of vitamin D is a mediator variable.
A moderator variable, on the other hand, affects the strength of the relationship between the dependent and the independent variables. In our example, if the exposure to sunlight reduced depression symptoms more for younger individuals than older individuals, then age is a moderator variable.