Research Design
A critical aspect of research is research design. If theres a flaw or two in the way youve conducted your research, all of your work has gone to waste and the conclusions you draw are null and void. Science is critical by naturehypotheses and theories constantly change as new ideas are discovered and old ideas are debunked. Scientists persistently question the works of others and they should evaluate their own. If youre familiar with research, you know that something is always bound to go wrong. And when it does, youll either have to start all over or youll come up with a faulty end result that critics will jump upon without mercy. So you see, research design is key. There are 3 types of research design: observational, correlational, and experimental.
Observational studies involve observation, obviously. It involves observing and collecting data without interference. The advantage of observational research design is that its as close to the real world as you can get, but the problem is that you cant control anything thats going on, so its difficult to come up with reliable data. You might be observing whether birds like to sit on a certain telephone wire, but maybe the weather wasnt ideal that day, or maybe theyve migrated, or maybe they just dont like that wire. There are a million different things that can interfere with your findings.
Correlational studies are a popular form of research and this is the kind of research design behind studies that tell you about associations, like your oatmeal canister telling you that eating oatmeal is associated with lower rates of heart disease. Correlational research is especially useful when researching things you cant control due to things like difficulty or ethics. For example, it seems unethical to make people smoke to see if they develop cancer, so all you can do is see whether people who are smoking are developing more cancer. This points to the disadvantage of correlationyou can show that things are related, but you cant show causality. This is why some people are able to say that smoking doesnt cause cancer, scientifically speaking. Also, some might say that oatmeal doesnt lower heart disease, because maybe people who choose to eat oatmeal just tend to be more health-conscious.
Experimental research design is the kind that can show causality, and provide the strongest scientific evidence. If you got a bunch of people and put them on a specific oatmeal diet, and then found that they had lower heart disease, then this would prove that oatmeal really is amazing. But note prove, because experiments have their limitations too. Experiments are the best controlled research design, but theyre the least realistic. People dont just get together and eat oatmeal in specific increments in the real world. Also, you have to consider the experiment participantsif you happened to pick a bunch of healthy people, then it might have inflated the results. Its necessary to gather a variety of participants so you can gather more accurate data that applies to a diverse population.
