Keep in mind that you’ll want to make sure that you specify a hypothesis (typically of the form, if X changes, then Y changes also). Note that experimental designs are built to identify *causal* inferences between two variables–here, we’ll call them X & Y. You’ll want to make sure you specify your independent variable that is controlled by the experimenter (X), and this is the hypothesized cause of the dependent variable (Y), which is measured in some quantitative fashion.
Then, you’ll want to include *random assignment* of participants in your study to levels of that independent variable. For instance, if you hypothesize that exposure to heat increases aggressive behavior, then you would randomly assign participants to experience either, say 70 or 90 degree temperatures in a room. This is a manipulation of the independent variable.
Then, you would measure aggressive behavior somehow–say, the number of swear words used in a conversation in the room.
The idea is that if random assignment is executed, then the only factor that differentiates the two conditions will be the independent variable (here, temperature), and that would be the cause of the change in the dependent variable (here, aggression).
Of course, you would want to use this basic structure to address a research question in your own area of interest…. You might think of examples of different designs, and how they could be applied to help understand the distinctions amongst these…
As always, I also encourage you to tie in the discussions to assignment content (your paper!!) as much as possible.
Imagine you were conducting research on the relationship between academic performance (e.g., better grades) and different types of music while studying.
1. How would you design the study if it were correlational?
2. How would you design the study if it were descriptive?
3. How would you design the study if it were experimental?