LITMIR.BIZ

      To test a hypothesis that exercise affects depression, we could have one group exercise and another not exercise.

      © iStockphoto.com/PeopleImages

      The idea that watching violence on television or playing video games with violent content increases aggression is certainly a reasonable and potentially important notion. The popular press has often tried to link violent video games with those who carry out school shootings. Yet before we can test this idea, we must define exactly what is meant by violence on television or in video games. Is a program with an unseen murder more violent than an exciting boxing match? Should violence be rated by how many minutes it appears on the screen, by the particular type of act, by how much blood is shown, or by a combination of all three? Likewise, to perform this research we need to devise some measure of aggression. We would have to adopt operational definitions.

      experimental group: a group that receives the independent variable (IV) in a study using the experimental method

      control group: in a research experiment, the group that is treated exactly like the experimental group except for not experiencing the independent variable (IV) being studied

      operational definition: a definition that presents a construct in terms of observable operation that can be measured and utilized in research

      independent variable (IV): the manipulated variable in an experimental study

      dependent variable (DV): in an experimental study, the variable that is said to depend on the action of another variable (the independent variable (IV))

      validity: truth of the results and capability of being supported

      confounding variables: unintended factors not chosen by the experimenter, but which influence the independent variable (IV)

      covary: the degree to which variables are related to one another

      An operational definition takes a general concept, such as depression or aggression or effectiveness, and places it within a given context. That is, it redefines the concept in terms of clearly observable operations that anyone can see and repeat. For example, we might define aggression as the number of times a child hits a toy after watching a violent video.

      In an experimental study, we want to know how one variable that we manipulate affects another variable. Think of a study in which you want to know if drinking coffee improves your memory. Whether or not someone drank coffee is the manipulated variable. This is also called the independent variable (IV). Memory in this example is the variable influenced by the coffee and is called the dependent variable (DV). That is, it depends on or is influenced by the IV.

      What other factors could influence a memory test? If we suspect that some unintended factor may also be operating, then the truth or validity of the experiment is seriously threatened. Thus, the conclusion that the IV influenced the DV could be questioned. In the memory experiment, if the control group was run in the morning and the experimental group in the afternoon, then time of day could have an effect. Whenever two or more IVs are operating, the unintended IVs (those not chosen by the experimenter) are called confounding variables.

      Other confounding variables may covary with the IV and be more difficult to notice. For example, assume that a researcher compared a new medication against a problem-solving approach for the treatment of anxiety. If she found the problem-solving approach led to a greater reduction in anxiety, could she conclude that problem solving produced the reduction? Although that is one possibility, it also may have been the case that spending time with a professional produced the reduction in anxiety. That is, because giving medications requires less time with a patient than discussing problem-solving techniques, the results found may not have been due to the IV as planned in the study but rather to a confounding variable of time with the patient.

      Does Playing Music Change the Brain?

      Let’s examine a specific study in which both experimental and correlational procedures were used. Thomas Elbert and his colleagues (Elbert, Pantev, Weinbruch, Rockstroh, & Taub, 1995) began with the idea that experience could change the manner in which connections in the brain were established. What these researchers needed to do was find a task that would allow them to measure change and make logical inferences. Since they were interested in long-term changes, they sought to find a skill that people learn in childhood. They decided on an experiment that involved playing a musical instrument.

      What musical instrument would you choose for such an experiment—piano, saxophone, violin, or another instrument? They chose the violin. Here is where logic and experimental design came in. By choosing the violin, these researchers were able to compare the differences between the violinists’ left and right hands and their representation in the brain. Violinists use their left hand to continuously finger the strings. The right hand moves the bow back and forth and does not require the same fine motor skills. If playing a violin for 20 years would affect the brain, then it should affect those areas involved with the left hand in a different manner from those involved with the right. This is exactly what they found.