Introduction to personality  (Page 15/21)

Since the investigator does not control the variables in correlational research, it is not possible to determine whether or not one variable causes the relationship. In the example used above, it certainly seems that studying more would lead to getting a better grade on a test. But consider another example: can money buy happiness? There is some evidence that wealthy people are happier than the average person, and that people in wealthy countries are happier than those in poorer countries. But does the money affect happiness? Certainly a million dollars in cash wouldn’t help much if you were stranded on a desert island, so what can it do for you at home? People with money can live in nicer, safer communities, they have access to better health care (so they may feel better physically), they may have more time to spend with their family and friends, and so in many ways their lives might be different. We can also look at the correlation the other way around; maybe happy people get more money. If you ran a company, and were going to hire or promote someone, wouldn’t you want to find someone who is friendly and outgoing? Wouldn’t you look for someone who other people will enjoy working with? So, maybe happy people do find it easier to be successful financially. Either way, we simply can’t be sure about which variable influences the other, or even if they influence each other at all. In order to do that, we must pursue experimental research.

In these figures, adapted from research conducted by the author (Kelland et al., 1989), we see two correlations reported in an actual study. In the figure on the left, we can see a significant positive correlation between the firing rate of dopamine neurons in the rat brain and the dose of the drug quinpirole needed to inhibit those cells. In the figure on the right, we can see that the correlation is eliminated (the dose of quinpirole needed is not related to the firing rate of the cell) following administration of the drug MDMA (more commonly known as Ecstasy!).

Experimental and Quasi-Experimental Designs

The experimental design is usually preferred within psychology, as with any other science. The goal is to control every aspect of the experiment and then manipulate a single variable, thus allowing us to attribute the results to that single manipulation. As a result, experiments allow us to make cause-and-effect statements about the relationships between the variables.

A simple experiment begins with defining the independent variable , the factor that will be manipulated, and the dependent variable , the factor that will be measured. Ideally, we then select our subjects in a random fashion, and assign them randomly to a control group and an experimental group . The experimental group is then exposed to the independent variable, whereas the control group is not. If we have successfully controlled all other variables through random selection of subjects (i.e., all subjects in a specified population have an equal chance of being selected for the study) and random assignment to the control and experimental groups (so that hopefully each group has an equal representation of gender, races, age, intelligence, personal habits, etc.), we should see a difference in the dependent variable that was caused by the independent variable.