The scientific method is a set of guidelines for conducting effective scientific research. It is a formula for gathering data, reaching conclusions, and establishing scientific theories. A scientific theory is a principle or idea that explains why something happens. Much of psychology is grounded in scientific theories, which in turn are created via the scientific method.
Steps of the Scientific Method
The scientific method typically follows these steps.
- Identify a research question and create a hypothesis. A hypothesis is a testable explanation for a phenomenon.
- Design and perform an experiment.
- Evaluate the original hypothesis by analyzing data from the experiment.
- Form conclusions about whether to accept or reject your hypothesis. If you believe your hypothesis was wrong, go back to step 1 and create a new one.
How to Write a Good Hypothesis
Many hypotheses take the following form: “If X is true, then Y will happen.” For example, a researcher could write, “If sleep deprivation makes a person forgetful, then people who don’t sleep the night before will get lower scores on a memory test than those who get 8 hours of sleep.”
A good hypothesis is testable, meaning each variable can be observed and measured. In the example above, a researcher could easily count how many hours each participant slept or calculate the score they got on a memory test. These measurements will result in objective numbers.
Subjective values, such as “love” or “evil”, are harder to measure. For instance, take the hypothesis, “If beautiful people are happy, then becoming beautiful will make a person happier.” Concepts such as “happy” and “beautiful” don’t have a universal definition. What one person describes as “content” or “plain,” another might call “joyful” or “pretty”.
So how do psychology studies measure emotions, opinions, and other subjective things? In many cases, scientists ask a person to report their subjective experience of the concept. For instance, a person could rate their mood on a 1-10 scale several times a week. By comparing later scores to the original score, a researcher can tell when a person’s mood improves or gets worse.
What Is a Control Group?
Most scientific experiments have what is called a control group. To explain this term and why it’s important, let’s pretend you are setting up a hypothetical drug experiment. Your hypothesis would be, “If Drug X treats anxiety, then people who have anxiety and take Drug X should stop having symptoms.”
On the surface, designing the experiment should be easy. Simply give Drug X to the study group and then record what happens. But how do you know the drug caused the improvements? What if some other variable treated the anxiety, like the participants getting more sleep during the study?
That is where the control group comes in. The control group is the default you compare the study group to. Ideally everything else about the two groups is the same: their ages, sleep schedules, food intake, etc. The only difference between the groups is whether they are taking Drug X.
In general, there are three possible scenarios for a control group and study group:
- If both groups see improvements, then some other variable may be contributing to the improvement. This doesn’t necessarily mean Drug X is useless. It simply tells you that other variables are more important than you realized.
- If the study group sees improvements but the control group does not, then Drug X is likely the cause of the change.
- If neither group sees improvements, then Drug X probably doesn’t work. This does not mean you had a bad hypothesis or even a bad experiment. After all, now you know that Drug X doesn’t work, rather than simply guessing. It may not be the result you wanted, but you still have more knowledge than you did before, which is the whole point of science.
In drug tests like this one, you may give the control group a placebo: a “fake” pill without any medicine inside. Sometimes people who believe a pill will treat their symptoms temporarily see improvements regardless of what’s actually in the pill. This is called the placebo effect. If the control group sees similar improvements as the study group, then the placebo effect is probably causing the change.
Researcher bias can also affect the results. For example, a researcher who worked really hard to create Drug X would likely want it to succeed. Even if they try to stay impartial, they are more likely to “see” signs of improvement. They might also subconsciously discourage participants from reporting anxiety symptoms through body language or tone of voice.
To guard against bias, you can hold a double-blind study. In other words, neither the study participants nor the researchers observing them would know which group received the real drug and which group received the placebo. This way, the researchers’ expectations would be less likely to color their perceptions.
Types of Studies
Although researchers tend to prefer creating double-blind studies, they are not always possible to run. Some fields of research make it difficult to establish rigorous controls.
For example, in a study of the effects of rape on victims, it would be unethical to set up a control group who is not raped and a study group that is. Researchers would have to look at effects after the fact. They might try to find groups that are similar in age, ethnicity, class, and other variables. However, there would be no way to find victims who grew up in exactly the same environments.
In other cases, a control group isn’t used in the experiment at all. Other psychological research methods include:
- Case Studies: These in-depth studies examine a person or a few people with a particular condition or life history. They can provide lots of information about individuals, but the results are not necessarily applicable to the general population. They may be useful when studying rare conditions.
- Longitudinal Studies: These studies look at a cohort of similarly-situated people over long periods of time. For instance, a study may follow traumatized veterans for ten years to see how their PTSD symptoms change over time.
- Survey Research: To be scientifically valid, surveys must use a representative sample of the study population. While surveys are often cheaper than other research methods, they do pose several problems. Respondents may lie in their responses in order to make themselves look good or to please the researchers. Respondents could also have limited self-awareness.
- Meta-analysis: This is a study of other studies. A meta-analysis looks at the results of similar studies and compares them to make broader conclusions about the data. For example, a meta-analysis could look at a dozen studies on divorce and report that most of the studies found communication issues to be a factor.
- American Psychological Association. APA concise dictionary of psychology. Washington, DC: American Psychological Association, 2009. Print.
- Colman, A. M. (2006). Oxford dictionary of psychology. New York, NY: Oxford University Press. Print.
- Helmenstine, A. M. (2018, September 2). 6 steps of the scientific method. Retrieved from https://www.thoughtco.com/steps-of-the-scientific-method-p2-606045
- Zimmermann, K. (2012, July 10). What is a Scientific Hypothesis? | Definition of Hypothesis. Retrieved from http://www.livescience.com/21490-what-is-a-scientific-hypothesis-definition-of-hypothesis.html
Last Updated: 05-20-2019
Leave a Comment
By commenting you acknowledge acceptance of GoodTherapy.org's Terms and Conditions of Use.