2-1 AND 2-2

The Building Blocks of Research Design

Putting the Scientific Method Into Action: Research Design

In Module One, the focus was on the scientific method and why it is such an important tool for learning about the world. The rest of this course will focus on how psychologists actually put the scientific method into action by designing research studies. After observing the world and reading academic literature, psychologists begin to put the scientific method into practice by posing a research question. This question leads to a hypothesis, which is a specific prediction about the answer to the research question. Then, since empirical data is at the heart of the scientific method, the psychologist will engage in research procedures in order to collect data that could potentially support the hypothesis.

As a concrete example, in this module you will read the beginning of an article about how the taste that is present in one’s mouth can potentially influence one’s moral judgments (Eskine, Kacinik, & Prinz, 2011). In it, you will see how the authors first pose a general question that is suggested by previous research on moral judgments. Then, they clearly articulate a hypothesis and outline their method to collect empirical data. These authors are putting the scientific method into action!

The Key Building Blocks of Research Design: Hypotheses

The formulation of a hypothesis is a critical step in the scientific method. In other words, psychologists do not conduct research aimlessly, or ―just to see what they might find.‖ Rather, psychologists read about a topic in advance and then use what they learn about the topic to develop not only a research question, but also a hypothesis about that question.

However, not all hypotheses are created equally. One of the critical principles of psychological research and science in general is that hypotheses must be testable and falsifiable. This means that a hypothesis has to be something that the researcher can actually test, and it has to be something that can possibly be proven wrong. As an example, consider the hypothesis ―All of our dreams have an important meaning.‖ How might a researcher collect data to answer this question? Is it even possible to answer this question? Not really—this is an example of a hypothesis that is not falsifiable. What evidence could we ever gather to show that a specific dream does not have an important meaning? It seems that no matter what, there could always be some interpretation of the dream that gives it a deeper meaning. From a scientific point of a view, such a hypothesis is essentially worthless. On the contrary, consider the hypothesis ―Dreaming is most common in the REM stage of sleep.‖ Now here is a hypothesis that is falsifiable—it is possible to collect data to disprove this hypothesis. A hypothesis such as this has scientific value because when it is tested, we learn something interesting about the world—either dreaming is most common in the REM stage of sleep or it is not.

The Key Building Blocks of Research Design: Variables

When developing a testable hypothesis to design a research study, it is important to note that we are really using the hypothesis to make a statement about variables. Anything that can vary from person to person or thing to thing is known as a variable. Take for instance the hypothesis ―Individuals in a bad mood are more likely than individuals in a good mood to act aggressively.‖ This hypothesis centers on the relationship between two key variables: mood and aggression. Data could be collected to test this hypothesis, and the data would either support or not support the hypothesis.

However, notice that to actually collect data, we first have to define our variables, bad mood and aggression. In other words, how exactly would we know if someone is in a bad mood, and how would we know if someone is being aggressive? This definition of the variable in the terms of the actual research being conducted is known as the operational definition. In this case, a bad mood could be operationally defined as scoring below a certain mark on a well-established measure of positive and negative mood. Aggression could be operationally defined as yelling at someone in a social situation. Note that no matter which variables are being studied, there are many different possible ways of coming up with an operational definition!

To go a step further, we must recognize that these variables fulfill different functions within research designs. When discussing variables in the context of research design, it is necessary to understand the distinction between independent and dependent variables. Independent variables refer to variables that a researcher manipulates in order to see their effect on the outcome of interest, the dependent variable. In the example above, a researcher might manipulate participants’ mood (by having them watch either a funny or sad video, perhaps) and then measure how aggressive they are in a social situation. Mood would be the independent variable because it was manipulated, whereas aggression would be the dependent variable because it was the outcome of interest. You can think of it as the dependent variable ―depends‖ on the independent variable. Of course, research situations can be a lot more complicated than simply having one independent variable and one dependent variable. A researcher may manipulate several different variables in a study, and they could certainly be interested in more than just one outcome.

Manipulating and Measuring Psychological Variables

To fully put the scientific method into action, it is not enough just to identify the variables relevant to one’s hypothesis. Ultimately, the key variables in a study have to be measured and/or manipulated! There are a variety of different, creative ways to manipulate variables in the research laboratory. For instance, in the article on taste and moral judgments, taste is manipulated by having some participants consume a bitter beverage, whereas others consume a sweet or neutral beverage (Eskine, Kacinik, & Prinz, 2011). In the example about mood and aggression, mood could be manipulated by having participants watch a funny or sad video. Alternatively, mood could be manipulated by having some participants recall a happy experience, whereas others recall a sad experience. The point is, there are many different ways to manipulate psychological variables!

There are also a multitude of ways to measure psychological variables. For instance, questionnaires can be used that ask people to answer questions about their behavior or their personality. Actual behavior can be observed and recorded. Further, there are many less obvious types of measurements, such as implicit measures that purport to measure psychological variables in an unobtrusive manner. The strengths and weaknesses of various types of psychological measurement will be a recurring theme in the articles you read this semester and the various module topics.

References

Eskine, K. J., Kacinik, N. A., & Prinz, F. J. (2011). A bad taste in the mouth: Gustatory disgust influences moral judgment. Psychological Science, 22(3), 295–299. Pixgood.com. (n.d.). Independent and dependent variables [Online image]. Retrieved from http://pixgood.com/independent-and-dependent-variable.html